Merge branch 'main' into codex/coupon-rate-exclude-abandoned

alembic/versions/comparison_below_minimum_as_success.py

@@ -1,76 +0,0 @@
-"""normalize granular comparison outcomes into terminal record statuses
-
-Revision ID: comparison_below_min_success
-Revises: limit_policy_global_bundle
-"""
-from __future__ import annotations
-
-from collections.abc import Sequence
-
-import sqlalchemy as sa
-from sqlalchemy.dialects import postgresql
-
-from alembic import op
-
-revision: str = "comparison_below_min_success"
-down_revision: str | Sequence[str] | None = "limit_policy_global_bundle"
-branch_labels: str | Sequence[str] | None = None
-depends_on: str | Sequence[str] | None = None
-
-_JSON = sa.JSON().with_variant(postgresql.JSONB(), "postgresql")
-_FAILED_OUTCOMES = (
-    "store_closed",
-    "store_not_found",
-    "items_not_found",
-    "no_delivery",
-    "unsupported",
-)
-
-
-def _comparison_record() -> sa.TableClause:
-    return sa.table(
-        "comparison_record",
-        sa.column("status", sa.String(16)),
-        sa.column("fail_reason", sa.String(256)),
-        sa.column("raw_payload", _JSON),
-    )
-
-
-def upgrade() -> None:
-    comparison_record = _comparison_record()
-    op.execute(
-        comparison_record.update()
-        .where(comparison_record.c.status == "below_minimum")
-        .values(status="success", fail_reason=None)
-    )
-    op.execute(
-        comparison_record.update()
-        .where(comparison_record.c.status.in_(_FAILED_OUTCOMES))
-        .values(status="failed")
-    )
-
-
-def downgrade() -> None:
-    comparison_record = _comparison_record()
-    # The write paths deliberately preserve the granular outcome. Restore only
-    # rows that this change normalized, without touching ordinary successes.
-    raw_outcome = sa.func.coalesce(
-        comparison_record.c.raw_payload["record_status"].as_string(),
-        comparison_record.c.raw_payload["status"].as_string(),
-    )
-    op.execute(
-        comparison_record.update()
-        .where(
-            comparison_record.c.status == "success",
-            raw_outcome == "below_minimum",
-        )
-        .values(status="below_minimum")
-    )
-    op.execute(
-        comparison_record.update()
-        .where(
-            comparison_record.c.status == "failed",
-            raw_outcome.in_(_FAILED_OUTCOMES),
-        )
-        .values(status=raw_outcome)
-    )

app/admin/repositories/coupon_data.py

@@ -166,6 +166,18 @@ def _session_to_row(
     point_stats: dict | None = None,
 ) -> dict:
     """CouponSession ORM → 明细行 dict(主表「领券数据」与「用户全部领券」抽屉共用)。"""
+    # 中途退出可能发生在第一张券产生终态之前，此时没有逐券事件。
+    # 明确返回 0/0，让前端区分「退出前无单券结果」与其它状态的埋点缺失。
+    if point_stats is not None:
+        point_success_count = point_stats["succeeded"]
+        point_total_count = point_stats["tried"]
+    elif r.status == "abandoned":
+        point_success_count = 0
+        point_total_count = 0
+    else:
+        point_success_count = None
+        point_total_count = None
+    point_event_count = point_stats["events"] if point_stats is not None else 0
     return {
         "id": r.id,
         "trace_id": r.trace_id,
@@ -182,8 +194,9 @@ def _session_to_row(
         "app_env": r.app_env,
         "started_at": r.started_at,
         "claimed_count": r.claimed_count,
-        "point_success_count": point_stats["succeeded"] if point_stats else None,
-        "point_total_count": point_stats["tried"] if point_stats else None,
+        "point_success_count": point_success_count,
+        "point_total_count": point_total_count,
+        "point_event_count": point_event_count,
         "trace_url": r.trace_url,
         "ad_revenue_yuan": ad_revenue_yuan,
     }
@@ -194,21 +207,24 @@ def _point_scores_by_trace(db: Session, trace_ids: list[str]) -> dict[str, dict[
     if not trace_ids:
         return {}
     succeeded = func.sum(case((CouponClaimEvent.status.in_(_SLOT_OK), 1), else_=0))
+    tried = func.sum(case((CouponClaimEvent.status.in_(_SLOT_TRIED), 1), else_=0))
     rows = db.execute(
         select(
             CouponClaimEvent.trace_id,
             succeeded.label("succeeded"),
-            func.count().label("tried"),
-        )
-        .where(
-            CouponClaimEvent.trace_id.in_(trace_ids),
-            CouponClaimEvent.status.in_(_SLOT_TRIED),
+            tried.label("tried"),
+            func.count().label("events"),
         )
+        .where(CouponClaimEvent.trace_id.in_(trace_ids))
         .group_by(CouponClaimEvent.trace_id)
     ).all()
     return {
-        trace_id: {"succeeded": int(success_count or 0), "tried": int(tried or 0)}
-        for trace_id, success_count, tried in rows
+        trace_id: {
+            "succeeded": int(success_count or 0),
+            "tried": int(tried_count or 0),
+            "events": int(event_count or 0),
+        }
+        for trace_id, success_count, tried_count, event_count in rows
         if trace_id is not None
     }
 
@@ -400,12 +416,15 @@ def coupon_user_records(db: Session, *, user_id: int, limit: int = 100) -> dict:
     total = db.execute(
         select(func.count()).select_from(CouponSession).where(CouponSession.user_id == user_id)
     ).scalar_one()
-    rev_map = crud_ecpm.revenue_yuan_by_trace(db, [r.trace_id for r in rows])
+    trace_ids = [r.trace_id for r in rows]
+    rev_map = crud_ecpm.revenue_yuan_by_trace(db, trace_ids)
+    point_stats_map = _point_scores_by_trace(db, trace_ids)
     return {
         "items": [
             _session_to_row(
                 r,
                 ad_revenue_yuan=rev_map.get(r.trace_id, 0.0),
+                point_stats=point_stats_map.get(r.trace_id),
             )
             for r in rows
         ],

app/admin/repositories/queries.py

@@ -39,34 +39,6 @@ from app.repositories import activity, ad_ecpm
 # 折算成可提现现金时,非广告金币来源的排除集(广告单独统计、人工调整不算"赚取")
 _NON_TASK_BIZ_TYPES = ("reward_video", "feed_ad_reward", "admin_grant", "admin_deduct")
 
-# comparison_record historically persisted a few granular business outcomes as
-# top-level statuses. Admin filters and metrics expose lifecycle buckets while
-# retaining the raw values until the data migration has run everywhere.
-_COMPARISON_STATUS_ALIASES = {
-    "success": ("success", "below_minimum"),
-    "failed": (
-        "failed",
-        "store_closed",
-        "store_not_found",
-        "items_not_found",
-        "no_delivery",
-        "unsupported",
-    ),
-    "cancelled": ("cancelled",),
-    "running": ("running",),
-}
-_COMPARISON_SUCCESS_STATUSES = _COMPARISON_STATUS_ALIASES["success"]
-_COMPARISON_FAILED_STATUSES = _COMPARISON_STATUS_ALIASES["failed"]
-_COMPARISON_COMPLETED_STATUSES = (
-    *_COMPARISON_SUCCESS_STATUSES,
-    *_COMPARISON_FAILED_STATUSES,
-)
-
-
-def _comparison_status_condition(status: str):
-    values = _COMPARISON_STATUS_ALIASES.get(status, (status,))
-    return ComparisonRecord.status.in_(values)
-
 # 信息流点位场景 → 金币记录「赚取途径」展示名;NULL/未知 = 历史未分类。
 _FEED_SCENE_LABEL = {
     "comparison": "比价信息流",
@@ -362,7 +334,7 @@ def _comparison_conditions(
             )
         )
     if status:
-        conditions.append(_comparison_status_condition(status))
+        conditions.append(ComparisonRecord.status == status)
     if business_type:
         conditions.append(ComparisonRecord.business_type == business_type)
     if store:
@@ -447,7 +419,7 @@ def _comparison_duration_aggregate_stmt(conditions: list, status: str, quantiles
         ),
     ).where(
         *conditions,
-        _comparison_status_condition(status),
+        ComparisonRecord.status == status,
         ComparisonRecord.total_ms.is_not(None),
     )
 
@@ -472,7 +444,7 @@ def _comparison_duration_aggregates(
             select(ComparisonRecord.total_ms)
             .where(
                 *conditions,
-                _comparison_status_condition(status),
+                ComparisonRecord.status == status,
                 ComparisonRecord.total_ms.is_not(None),
             )
             .order_by(ComparisonRecord.total_ms)
@@ -502,11 +474,11 @@ def comparison_records_summary(
     row = db.execute(
         select(
             func.count(ComparisonRecord.id),
-            func.sum(case((ComparisonRecord.status.in_(_COMPARISON_COMPLETED_STATUSES), 1), else_=0)),
-            func.sum(case((ComparisonRecord.status.in_(_COMPARISON_SUCCESS_STATUSES), 1), else_=0)),
+            func.sum(case((ComparisonRecord.status.in_(("success", "failed")), 1), else_=0)),
+            func.sum(case((ComparisonRecord.status == "success", 1), else_=0)),
             func.avg(ComparisonRecord.llm_cost_yuan),
             func.sum(case((
-                ComparisonRecord.status.in_(_COMPARISON_SUCCESS_STATUSES)
+                (ComparisonRecord.status == "success")
                 & (ComparisonRecord.saved_amount_cents > 0), 1
             ), else_=0)),
             func.sum(case((ComparisonRecord.status == "cancelled", 1), else_=0)),

app/admin/repositories/stats.py

@@ -566,6 +566,10 @@ def dashboard_overview(
         )
     ).all()
     coupon_started = len(period_coupon_sessions)
+    coupon_abandoned = sum(s.status == "abandoned" for s in period_coupon_sessions)
+    # 用户主动中途退出不代表领券流程失败，不进入整场成功率样本。
+    # started / failed 仍留在分母：前者是尚未形成终态的流失，后者是实际执行失败。
+    coupon_success_denominator = coupon_started - coupon_abandoned
     coupon_completed_elapsed = sorted(
         s.elapsed_ms
         for s in period_coupon_sessions
@@ -731,9 +735,13 @@ def dashboard_overview(
             },
             "coupon": {
                 "started": coupon_started,
+                "abandoned": coupon_abandoned,
+                "success_denominator": coupon_success_denominator,
                 "all_success": coupon_all_success,
                 "success_rate": (
-                    round(coupon_all_success / coupon_started, 4) if coupon_started else None
+                    round(coupon_all_success / coupon_success_denominator, 4)
+                    if coupon_success_denominator
+                    else None
                 ),
                 "point_success": coupon_point_success,
                 "points_per_session": coupon_points_per_session,

app/admin/routers/comparison.py

@@ -35,7 +35,7 @@ def list_comparison_records(
     db: AdminDb,
     user_id: Annotated[int | None, Query()] = None,
     phone: Annotated[str | None, Query(description="手机号前缀")] = None,
-    status: Annotated[str | None, Query(pattern="^(success|failed|cancelled|running)$")] = None,
+    status: Annotated[str | None, Query(pattern="^(success|failed|cancelled)$")] = None,
     business_type: Annotated[str | None, Query()] = None,
     store: Annotated[str | None, Query(description="店名子串模糊匹配")] = None,
     product: Annotated[str | None, Query(description="商品名子串模糊匹配")] = None,
@@ -66,7 +66,7 @@ def comparison_records_summary(
     db: AdminDb,
     user_id: Annotated[int | None, Query()] = None,
     phone: Annotated[str | None, Query(description="手机号前缀")] = None,
-    status: Annotated[str | None, Query(pattern="^(success|failed|cancelled|running)$")] = None,
+    status: Annotated[str | None, Query(pattern="^(success|failed|cancelled)$")] = None,
     business_type: Annotated[str | None, Query()] = None,
     store: Annotated[str | None, Query(description="店名子串模糊匹配")] = None,
     product: Annotated[str | None, Query(description="商品名子串模糊匹配")] = None,

app/admin/schemas/comparison.py

@@ -20,7 +20,7 @@ class AdminComparisonListItem(BaseModel):
     trace_id: str
     # admin 是 debug 工具,无条件下发 trace_url(不看 user.debug_trace_enabled)
     trace_url: str | None = None
-    status: str  # success / failed / cancelled / running;旧细分值由前端兼容映射
+    status: str
     information: str | None = None
     store_name: str | None = None
     product_names: str | None = None   # 下单商品名派生串(顿号分隔;「商品」列展示 + 商品搜索)
@@ -83,7 +83,6 @@ class AdminComparisonDetail(AdminComparisonListItem):
     skipped_dish_count: int | None = None
     device_id: str | None = None
     items: list = []
-    platforms: list = []  # pricebot 渲染就绪的逐平台卡片模型(status=ok/业务失败细分)
     comparison_results: list = []   # 逐平台对比(价格/rank/coupon/打烊...)
     skipped_dish_names: list = []
     # 全量环境

app/admin/schemas/coupon_data.py

@@ -79,10 +79,16 @@ class CouponDataRow(BaseModel):
     started_at: datetime = Field(..., description="发起时刻(明细「时间」列)")
     claimed_count: int | None = None
     point_success_count: int | None = Field(
-        None, description="本次成功单券数(success+already_claimed);无逐券事件为空"
+        None,
+        description="本次成功单券数(success+already_claimed);中途退出且无逐券结果为0,其它无事件为空",
     )
     point_total_count: int | None = Field(
-        None, description="本次尝试单券数(success+already_claimed+failed,不含 skipped);无逐券事件为空"
+        None,
+        description="本次尝试单券数(success+already_claimed+failed,不含 skipped);中途退出且无逐券结果为0,其它无事件为空",
+    )
+    point_event_count: int = Field(
+        0,
+        description="本次全部逐券事件数(含 skipped);用于区分无有效计分事件与完全无事件",
     )
     trace_url: str | None = Field(None, description="pricebot 公网 trace 链接(仅 completed 有);admin 渲染可点链接,无则显示可复制 trace_id")
     ad_revenue_yuan: float = Field(

app/admin/schemas/dashboard.py

@@ -70,6 +70,9 @@ class DashboardPeriodCoupon(BaseModel):
     成功口径 success+already_claimed(与「我的」页累计领券一致)。"""
 
     started: int = 0
+    # 用户主动中途退出，不计入整场成功率分母。
+    abandoned: int = 0
+    success_denominator: int = 0
     # 全部领成功的次数:completed 且当日该设备全部点位成功
     all_success: int = 0
     success_rate: float | None = None

app/api/v1/feedback.py

@@ -54,13 +54,11 @@ def _app_status(db_status: str) -> str:
 
 
 def _record_out(fb) -> FeedbackRecordOut:
-    images = fb.images or []
     return FeedbackRecordOut(
         id=fb.id,
         content=fb.content,
         scene=getattr(fb, "scene", None),
-        images=images,
-        image_thumbnails=[media.feedback_thumbnail_url(url) for url in images],
+        images=fb.images or [],
         status=_app_status(fb.status),
         reject_reason=getattr(fb, "reject_reason", None),
         reward_coins=getattr(fb, "reward_coins", None),
@@ -86,7 +84,7 @@ async def submit_feedback(
     device_model: str = Form(default=""),
     rom_name: str = Form(default=""),
     android_version: str = Form(default=""),
-    images: list[UploadFile] = File(default=[]),  # noqa: B008 - FastAPI dependency declaration
+    images: list[UploadFile] = File(default=[]),
 ) -> FeedbackOut:
     content = content.strip()
     contact = contact.strip()

app/core/media.py

@@ -11,7 +11,6 @@
 from __future__ import annotations
 
 import json
-import logging
 import os
 import secrets
 import subprocess
@@ -19,17 +18,8 @@ from decimal import ROUND_HALF_UP, Decimal, InvalidOperation
 from pathlib import Path
 from typing import Any
 
-from PIL import Image, ImageOps
-
 from app.core.config import settings
 
-logger = logging.getLogger("shagua.media")
-
-_FEEDBACK_DIR = "feedback"
-_FEEDBACK_THUMB_DIR = "feedback_thumbs"
-_FEEDBACK_THUMB_MAX_PX = 256
-_FEEDBACK_THUMB_QUALITY = 78
-
 
 class MediaError(Exception):
     """上传文件不合法(类型/大小)。调用方转 400。"""
@@ -78,86 +68,8 @@ def save_avatar(user_id: int, data: bytes) -> str:
 
 
 def save_feedback_image(user_id: int, data: bytes) -> str:
-    """保存反馈截图并预生成历史页缩略图，返回原图相对 URL。"""
-    url = _save_image(_FEEDBACK_DIR, user_id, data)
-    # 缩略图失败不影响反馈受理；读取缩略图 URL 时会按需重试并回退原图。
-    ensure_feedback_thumbnail(url)
-    return url
-
-
-def feedback_thumbnail_url(image_url: str) -> str:
-    """把反馈原图 URL 映射成确定的缩略图 URL，不在 records 接口内做图片解码。
-
-    上传文件名由服务端生成且不会覆盖；旧数据在客户端真正请求可见图片时按需补图。
-    """
-    paths = _feedback_thumbnail_paths(image_url)
-    return paths[2] if paths is not None else image_url
-
-
-def _feedback_thumbnail_paths(image_url: str) -> tuple[Path, Path, str] | None:
-    prefix = f"{settings.MEDIA_URL_PREFIX}/{_FEEDBACK_DIR}/"
-    if not image_url.startswith(prefix):
-        return None
-
-    filename = image_url.removeprefix(prefix)
-    # 只接受当前目录下的单个文件名，避免数据库脏数据造成路径穿越。
-    if not filename or Path(filename).name != filename:
-        return None
-
-    source = _media_dir(_FEEDBACK_DIR) / filename
-    thumb_name = f"{Path(filename).stem}.jpg"
-    thumb = _media_dir(_FEEDBACK_THUMB_DIR) / thumb_name
-    thumb_url = f"{settings.MEDIA_URL_PREFIX}/{_FEEDBACK_THUMB_DIR}/{thumb_name}"
-    return source, thumb, thumb_url
-
-
-def ensure_feedback_thumbnail(image_url: str) -> Path | None:
-    """确保缩略图存在并返回文件；生成失败时回退原图，供动态缩略图路由使用。"""
-    paths = _feedback_thumbnail_paths(image_url)
-    if paths is None:
-        return None
-    source, thumb, _ = paths
-    if thumb.is_file():
-        return thumb
-    if not source.is_file():
-        return None
-
-    temp = thumb.with_name(f".{thumb.name}.{secrets.token_hex(4)}.tmp")
-    try:
-        with Image.open(source) as opened:
-            image = ImageOps.exif_transpose(opened)
-            image.thumbnail(
-                (_FEEDBACK_THUMB_MAX_PX, _FEEDBACK_THUMB_MAX_PX),
-                Image.Resampling.LANCZOS,
-            )
-            if image.mode != "RGB":
-                image = image.convert("RGB")
-            image.save(
-                temp,
-                format="JPEG",
-                quality=_FEEDBACK_THUMB_QUALITY,
-                optimize=True,
-            )
-        os.replace(temp, thumb)
-        return thumb
-    except (Image.DecompressionBombError, OSError, ValueError):
-        logger.warning("生成反馈缩略图失败: %s", source, exc_info=True)
-        return source
-    finally:
-        temp.unlink(missing_ok=True)
-
-
-def feedback_thumbnail_file(filename: str) -> Path | None:
-    """由缩略图文件名找到原反馈图并按需生成，非法/不存在返回 None。"""
-    if not filename or Path(filename).name != filename or Path(filename).suffix.lower() != ".jpg":
-        return None
-    stem = Path(filename).stem
-    for ext in (".jpg", ".png", ".webp"):
-        original = _media_dir(_FEEDBACK_DIR) / f"{stem}{ext}"
-        if original.is_file():
-            original_url = f"{settings.MEDIA_URL_PREFIX}/{_FEEDBACK_DIR}/{original.name}"
-            return ensure_feedback_thumbnail(original_url)
-    return None
+    """保存反馈截图,返回相对 URL(`/media/feedback/<file>`)。"""
+    return _save_image("feedback", user_id, data)
 
 
 def save_report_image(user_id: int, data: bytes) -> str:

app/main.py

@@ -10,7 +10,7 @@ from collections.abc import AsyncIterator
 from contextlib import asynccontextmanager
 from pathlib import Path
 
-from fastapi import FastAPI, HTTPException
+from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import FileResponse
 from fastapi.staticfiles import StaticFiles
@@ -44,7 +44,6 @@ from app.api.v1.tasks import router as tasks_router
 from app.api.v1.user import router as user_router
 from app.api.v1.wallet import router as wallet_router
 from app.api.v1.wxpay import router as wxpay_router
-from app.core import media
 from app.core.config import settings
 from app.core.cps_reconcile_worker import (
     start_cps_reconcile_worker,
@@ -83,19 +82,6 @@ setup_logging(debug=settings.APP_DEBUG)
 logger = logging.getLogger("shagua.main")
 
 
-class FeedbackMediaStaticFiles(StaticFiles):
-    """反馈原图/缩略图文件名不可变，可长期缓存，避免列表反复回源。"""
-
-    async def get_response(self, path: str, scope):
-        response = await super().get_response(path, scope)
-        media_path = path.replace("\\", "/").lstrip("/")
-        if response.status_code == 200 and media_path.startswith(
-            ("feedback/", "feedback_thumbs/")
-        ):
-            response.headers["Cache-Control"] = "public, max-age=31536000, immutable"
-        return response
-
-
 @asynccontextmanager
 async def lifespan(_: FastAPI) -> AsyncIterator[None]:
     # 提示而非强制建表:生产用 alembic upgrade head,本地 dev 也建议先跑一次 migration。
@@ -226,24 +212,8 @@ def download_apk() -> FileResponse:
     )
 
 
-@app.get(
-    f"{settings.MEDIA_URL_PREFIX}/feedback_thumbs/{{filename}}",
-    tags=["feedback"],
-    include_in_schema=False,
-)
-def feedback_thumbnail(filename: str) -> FileResponse:
-    """旧反馈图按首次可见请求补缩略图；新图上传时已预生成。"""
-    path = media.feedback_thumbnail_file(filename)
-    if path is None:
-        raise HTTPException(status_code=404, detail="图片不存在")
-    return FileResponse(
-        path,
-        headers={"Cache-Control": "public, max-age=31536000, immutable"},
-    )
-
-
 app.mount(
     settings.MEDIA_URL_PREFIX,
-    FeedbackMediaStaticFiles(directory=str(_media_root)),
+    StaticFiles(directory=str(_media_root)),
     name="media",
 )

app/models/comparison.py

@@ -97,7 +97,7 @@ class ComparisonRecord(Base):
     total_dish_count: Mapped[int | None] = mapped_column(Integer, nullable=True)
     skipped_dish_count: Mapped[int | None] = mapped_column(Integer, nullable=True)
 
-    # success(流程正常完成,含 below_minimum)/ failed(技术异常或未形成可比报价,含店铺打烊等)
+    # success(拿到有效对比)/ failed(出错或没采到目标价)
     status: Mapped[str] = mapped_column(String(16), nullable=False, default="success")
     # done 帧 information 文案。成功:"在美团找到同店,到手价 ¥X…";
     # 失败:具体原因(如"美团、京东外卖均未找到该商品")。前端在比价失败时当原因展示。

app/repositories/comparison.py

@@ -76,30 +76,6 @@ _BIZ_STATUS_PRIORITY = (
 )
 
 
-def _normalize_record_status(status: str | None) -> str | None:
-    """Map granular business outcomes onto the record lifecycle status.
-
-    ``record_status`` describes the business outcome, while
-    ``comparison_record.status`` is also the completed-comparison flag used by
-    milestones, stats and idempotent rewards. ``below_minimum`` is a completed
-    success because the target cart produced a trustworthy conclusion. Other
-    known target-side outcomes did not produce a comparable quote and belong
-    to the failed record bucket. The granular outcome remains in ``raw_payload``
-    and ``platform_results`` for result rendering.
-    """
-    if status == "below_minimum":
-        return "success"
-    if status in {
-        "store_closed",
-        "store_not_found",
-        "items_not_found",
-        "no_delivery",
-        "unsupported",
-    }:
-        return "failed"
-    return status
-
-
 def _store_closed_text(reason: str | None) -> str:
     """打烊/暂停营业/休息类 reason 常带脏店名元数据 → 只留结论,套简短模板。"""
     r = reason or ""
@@ -188,10 +164,9 @@ def _derive(payload: ComparisonRecordIn) -> dict:
 
     is_source_best = best.is_source if best is not None else None
 
-    # status:优先 pricebot record_status → 客户端显式 status → 兜底派生。
-    # below_minimum 是已形成可信结论的正常完成态,记录级归 success;细分结局仍完整保留在
-    # raw_payload/platform_results,供结果卡展示"未满起送"。
-    status = _normalize_record_status(payload.record_status or payload.status)
+    # status:优先 pricebot record_status(区分 below_minimum/store_closed) → 客户端显式 status
+    # → 兜底"非源且有价"=success/否则 failed。record_status 让"未满起送"不再塌缩成 failed。
+    status = payload.record_status or payload.status
     if status is None:
         has_valid_target = any(
             (not r.is_source) and r.price is not None for r in results
@@ -226,9 +201,7 @@ def upsert_record(
     # 单源派生: 与 harvest_done 一致, payload 带 platforms 时从它派生(唯一真相源
     # _derive_from_platforms), 老客户端不带 platforms 时回退 _derive(从 comparison_results)。
     if payload.platforms:
-        derived = _derive_from_platforms(
-            payload.platforms, payload.record_status or payload.status
-        )
+        derived = _derive_from_platforms(payload.platforms, payload.record_status)
         # 对齐 _derive 返回键(#189 fail_reason): 两路径 fields 键集一致, 覆盖已有行时不残留旧值
         derived["fail_reason"] = (
             _derive_fail_display(payload.information, payload.platform_results or {})
@@ -383,10 +356,10 @@ def _derive_from_results(
         "saved_amount_cents": saved_amount_cents,
         "is_source_best": best.get("is_source") if best else None,
         "store_name": (src_row or {}).get("store_name") or None,
-        # below_minimum 已完成到购物车并形成可信结论,记录级计 success;细分结局仍在
-        # raw_payload/platform_results。旧 pricebot 未下发 record_status 时回退二态派生。
-        "status": _normalize_record_status(record_status)
-        or ("success" if has_valid_target else "failed"),
+        # 记录级结局: 优先用 pricebot 下发的 record_status(区分 below_minimum/store_closed,
+        # 不再把"未满起送"塌缩成 failed → 记录页不再误报"网络开小差"); 旧 pricebot 未下发时
+        # 回退老的 success/failed 二态派生, 向后兼容。
+        "status": record_status or ("success" if has_valid_target else "failed"),
     }
 
 
@@ -439,8 +412,7 @@ def _derive_from_platforms(
         "saved_amount_cents": saved_amount_cents,
         "is_source_best": (best.get("role") == "source") if best else None,
         "store_name": store_name or None,
-        "status": _normalize_record_status(record_status)
-        or ("success" if has_valid_target else "failed"),
+        "status": record_status or ("success" if has_valid_target else "failed"),
     }
 
 
@@ -612,8 +584,7 @@ def harvest_done(
     行不存在(理论上帧0已建;防御)则新建。"""
     results = done_params.get("comparison_results") or []
     # 展示模型统一数组(pricebot 新增, 每平台一行自带 status/is_best): 原样存, 记录页据此直渲染。
-    # record_status: 记录级业务结局(success/below_minimum/store_closed/failed)。其中
-    # below_minimum 是正常完成态,持久化 status 归 success,原值仍随 done_params 落 raw_payload。
+    # record_status: 记录级结局(success/below_minimum/store_closed/failed), 覆盖老二态派生。
     platforms = done_params.get("platforms") or []
     record_status = done_params.get("record_status")
     # 单源派生: platforms(含 pricebot 权威 is_best)是唯一真相源, best_*/source_*/saved/status

app/schemas/compare_record.py

@@ -111,9 +111,8 @@ class ComparisonRecordIn(BaseModel):
     # status/is_best/display/display_order,记录页据此直渲染。宽松 list[dict] 存(结构由
     # pricebot 定,server 只原样落库),前端读它、老记录空时回退 comparison_results。
     platforms: list[dict] = Field(default_factory=list)
-    # 记录级业务结局(pricebot 下发): success/below_minimum/store_closed/failed。
-    # below_minimum 表示流程正常完成,持久化主状态归 success;store_closed/items_not_found 等
-    # 已知无报价结局归 failed。原值仍随 raw_payload 落库,admin/记录页从 platform_results 展示细分结论。
+    # 记录级结局(pricebot 下发): success/below_minimum/store_closed/failed。让"未满起送"不再
+    # 被塌缩成 failed。_derive 优先用它、其次客户端 status、再兜底二态派生。
     record_status: str | None = None
     # 逐平台结局摘要(含失败平台的细分原因 status: store_not_found/items_not_found/below_minimum/
     # unsupported/...)。来自 done.params.platform_results,客户端透传;落 raw_payload(不单列),

app/schemas/feedback.py

@@ -33,8 +33,6 @@ class FeedbackRecordOut(BaseModel):
     # 比价反馈的问题场景(找错商品/优惠不对…);普通反馈为 None
     scene: str | None = None
     images: list[str] = Field(default_factory=list)
-    # 与 images 下标一一对应；生成失败时该项回退原图 URL，兼容历史数据。
-    image_thumbnails: list[str] = Field(default_factory=list)
     status: str
     reject_reason: str | None = None
     reward_coins: int | None = None

pyproject.toml

@@ -35,9 +35,6 @@ dependencies = [
     # multipart form (FastAPI 表单上传依赖)
     "python-multipart>=0.0.9",
 
-    # 用户反馈截图缩略图，避免 App 历史页为 48dp 小图下载数 MB 原图
-    "pillow>=11.0.0",
-
     # admin 后台账号密码 hash(用户侧是手机号+验证码登录,不需要密码;admin 才用)
     "bcrypt>=4.0.0",
 

tests/test_admin_read.py

@@ -1,7 +1,7 @@
 """Admin M2 读接口测试:大盘聚合 + 用户/流水/提现/反馈列表 + 鉴权拦截。"""
 from __future__ import annotations
 
-from datetime import UTC, datetime
+from datetime import UTC, date, datetime
 
 import pytest
 from fastapi.testclient import TestClient
@@ -12,6 +12,7 @@ from app.admin.repositories import admin_user as admin_repo
 from app.admin.repositories import queries
 from app.db.session import SessionLocal, engine
 from app.models.comparison import ComparisonRecord
+from app.models.coupon_state import CouponClaimRecord, CouponSession
 from app.models.feedback import Feedback
 from app.models.invite import InviteRelation
 from app.models.savings import SavingsRecord
@@ -120,6 +121,86 @@ def test_dashboard_period_comparison_is_aggregated_by_backend(
     assert comparison["token_cost_total_yuan"] == pytest.approx(1.0)
 
 
+def test_dashboard_coupon_success_rate_excludes_abandoned_sessions(
+    admin_client: TestClient, admin_token: str
+) -> None:
+    started_date = date(2038, 1, 16)
+    started_at = datetime(2038, 1, 16, 8, tzinfo=UTC)
+    sessions = [
+        ("coupon-rate-completed-1", "coupon-rate-device-1", "completed"),
+        ("coupon-rate-completed-2", "coupon-rate-device-2", "completed"),
+        ("coupon-rate-failed", "coupon-rate-device-3", "failed"),
+        ("coupon-rate-abandoned", "coupon-rate-device-4", "abandoned"),
+    ]
+    db = SessionLocal()
+    try:
+        for trace_id, device_id, status in sessions:
+            db.add(
+                CouponSession(
+                    trace_id=trace_id,
+                    device_id=device_id,
+                    status=status,
+                    app_env="prod",
+                    platforms=["meituan-waimai"],
+                    started_at=started_at,
+                    started_date=started_date,
+                )
+            )
+        for index, device_id in enumerate(("coupon-rate-device-1", "coupon-rate-device-2")):
+            db.add(
+                CouponClaimRecord(
+                    device_id=device_id,
+                    coupon_id=f"mt_dashboard_rate_{index}",
+                    claim_date=started_date,
+                    status="success",
+                    app_env="prod",
+                )
+            )
+        db.commit()
+    finally:
+        db.close()
+
+    response = admin_client.get(
+        "/admin/api/stats/overview",
+        params={"date_from": "2038-01-16", "date_to": "2038-01-16"},
+        headers=_auth(admin_token),
+    )
+    assert response.status_code == 200, response.text
+    coupon = response.json()["period"]["coupon"]
+    assert coupon["started"] == 4
+    assert coupon["abandoned"] == 1
+    assert coupon["success_denominator"] == 3
+    assert coupon["all_success"] == 2
+    assert coupon["success_rate"] == pytest.approx(2 / 3, abs=0.0001)
+
+    db = SessionLocal()
+    try:
+        db.add(
+            CouponSession(
+                trace_id="coupon-rate-only-abandoned",
+                device_id="coupon-rate-device-only-abandoned",
+                status="abandoned",
+                app_env="prod",
+                platforms=["meituan-waimai"],
+                started_at=datetime(2038, 1, 17, 8, tzinfo=UTC),
+                started_date=date(2038, 1, 17),
+            )
+        )
+        db.commit()
+    finally:
+        db.close()
+
+    empty_denominator_response = admin_client.get(
+        "/admin/api/stats/overview",
+        params={"date_from": "2038-01-17", "date_to": "2038-01-17"},
+        headers=_auth(admin_token),
+    )
+    assert empty_denominator_response.status_code == 200
+    only_abandoned = empty_denominator_response.json()["period"]["coupon"]
+    assert only_abandoned["success_denominator"] == 0
+    assert only_abandoned["success_rate"] is None
+
+
 def test_user_list_and_detail(admin_client: TestClient, admin_token: str) -> None:
     uid = _seed_user_with_data("13800000002")
     r = admin_client.get("/admin/api/users", headers=_auth(admin_token))
@@ -720,16 +801,6 @@ def test_comparison_records_show_readable_device_and_rom_version(
             rom_name="OriginOS",
             rom_version=4,
             android_version="14",
-            platforms=[
-                {
-                    "platform_id": "meituan",
-                    "platform_name": "美团",
-                    "status": "ok",
-                    "role": "target",
-                    "price": 18.8,
-                    "is_best": True,
-                }
-            ],
         )
         db.add(record)
         db.commit()
@@ -757,8 +828,6 @@ def test_comparison_records_show_readable_device_and_rom_version(
     assert detail.status_code == 200, detail.text
     assert detail.json()["device_model_name"] == "vivo Y77e"
     assert detail.json()["rom_version"] == 4
-    assert detail.json()["platforms"][0]["status"] == "ok"
-    assert detail.json()["platforms"][0]["is_best"] is True
 
 
 def test_comparison_records_show_real_order_status(

tests/test_compare_harvest.py

@@ -13,7 +13,6 @@ import uuid
 from unittest.mock import MagicMock, patch
 
 import httpx
-import pytest
 from sqlalchemy import select
 
 from app.db.session import SessionLocal
@@ -26,26 +25,6 @@ def _tid() -> str:
     return uuid.uuid4().hex
 
 
-@pytest.mark.parametrize(
-    ("raw_status", "record_status"),
-    [
-        ("success", "success"),
-        ("below_minimum", "success"),
-        ("failed", "failed"),
-        ("store_closed", "failed"),
-        ("store_not_found", "failed"),
-        ("items_not_found", "failed"),
-        ("no_delivery", "failed"),
-        ("unsupported", "failed"),
-        ("cancelled", "cancelled"),
-        ("running", "running"),
-        (None, None),
-    ],
-)
-def test_record_status_normalization(raw_status, record_status) -> None:
-    assert crud._normalize_record_status(raw_status) == record_status
-
-
 def _done_params() -> dict:
     """一份典型 done 帧 params:美团 25 元 vs 源淘宝闪购 30 元 → 省 5 元、success。"""
     return {
@@ -132,116 +111,6 @@ def test_harvest_done_derives_and_newly_success_once(client) -> None:
         assert newly2 is False
 
 
-def test_harvest_done_below_minimum_counts_as_completed_success(client) -> None:
-    """未达起送是可信业务结论:主状态/完成奖励归 success,细分结局仍留在 raw_payload。"""
-    tid = _tid()
-    done_below_minimum = {
-        "record_status": "below_minimum",
-        "comparison_results": [
-            {
-                "platform_id": "meituan",
-                "platform_name": "美团",
-                "package": "com.sankuai.meituan",
-                "price": 59.0,
-                "is_source": True,
-                "rank": 1,
-                "store_name": "测试店",
-                "items": [{"name": "红乌苏", "qty": 1}],
-            },
-        ],
-        "platform_results": {
-            "meituan": {"is_source": True, "status": "source", "price": 59.0},
-            "taobao_flash": {
-                "is_source": False,
-                "status": "below_minimum",
-                "reason": "购物车未达起送门槛（差 ¥25.2）",
-            },
-        },
-        "information": "淘宝未达起送门槛，可加菜凑单后下单",
-    }
-    with SessionLocal() as db:
-        crud.harvest_running(db, trace_id=tid, user_id=None)
-        rec, newly = crud.harvest_done(
-            db, trace_id=tid, user_id=None, done_params=done_below_minimum
-        )
-
-        assert rec.status == "success"
-        assert newly is True
-        assert rec.fail_reason is None
-        assert rec.raw_payload["record_status"] == "below_minimum"
-        assert (
-            rec.raw_payload["platform_results"]["taobao_flash"]["status"]
-            == "below_minimum"
-        )
-
-
-def test_harvest_done_platforms_below_minimum_counts_as_success(client) -> None:
-    """新 platforms 单源派生路径也必须执行同一 below_minimum → success 归一化。"""
-    tid = _tid()
-    with SessionLocal() as db:
-        rec, newly = crud.harvest_done(
-            db,
-            trace_id=tid,
-            user_id=None,
-            done_params={
-                "record_status": "below_minimum",
-                "platforms": [
-                    {
-                        "role": "source",
-                        "platform_id": "meituan",
-                        "platform_name": "美团",
-                        "price": 59.0,
-                        "store_name": "测试店",
-                        "items": [{"name": "红乌苏", "qty": 1}],
-                    },
-                    {
-                        "role": "target",
-                        "platform_id": "taobao_flash",
-                        "platform_name": "淘宝",
-                        "status": "below_minimum",
-                        "price": None,
-                    },
-                ],
-                "platform_results": {
-                    "taobao_flash": {
-                        "is_source": False,
-                        "status": "below_minimum",
-                        "reason": "购物车未达起送门槛",
-                    }
-                },
-            },
-        )
-
-        assert rec.status == "success"
-        assert newly is True
-        assert rec.raw_payload["record_status"] == "below_minimum"
-
-
-def test_legacy_upsert_below_minimum_counts_as_success(client) -> None:
-    """灰度期客户端直报路径无论走 status 还是 record_status 都不能落第四种主状态。"""
-    tid = _tid()
-    payload = ComparisonRecordIn(
-        trace_id=tid,
-        business_type="food",
-        status="below_minimum",
-        comparison_results=[],
-        platform_results={
-            "taobao_flash": {
-                "is_source": False,
-                "status": "below_minimum",
-                "reason": "购物车未达起送门槛",
-            }
-        },
-        information="淘宝未达起送门槛，可加菜凑单后下单",
-    )
-    with SessionLocal() as db:
-        rec = crud.upsert_record(db, user_id=987654, payload=payload)
-
-        assert rec.status == "success"
-        assert rec.fail_reason is None
-        assert rec.raw_payload["status"] == "below_minimum"
-
-
 def test_harvest_done_failed_derives_fail_reason(client) -> None:
     """failed 记录:记录级 information 笼统,但 fail_reason 从 platform_results 救出具体原因
     (id 3030 型:美团系统失败 + 京东 items_not_found → 展示京东那条)。"""

tests/test_comparison_admin_summary.py

@@ -23,7 +23,7 @@ def test_postgresql_duration_summary_uses_ordered_set_aggregates() -> None:
     )
 
     assert sql.count("percentile_cont") == 4
-    assert "comparison_record.status IN ('success', 'below_minimum')" in sql
+    assert "comparison_record.status = 'success'" in sql
 
 
 def test_summary_uses_only_success_durations_and_filters_beijing_date() -> None:
@@ -34,9 +34,6 @@ def test_summary_uses_only_success_durations_and_filters_beijing_date() -> None:
             ("summary-success-b", "success", 3000, 2.0, 0),
             ("summary-failed", "failed", 100_000, 3.0, 0),
             ("summary-cancelled", "cancelled", 5000, 4.0, 0),
-            ("summary-below-minimum", "below_minimum", 5000, None, 0),
-            ("summary-store-closed", "store_closed", 200_000, None, 0),
-            ("summary-running", "running", 4000, None, 0),
         ]
         for trace_id, status, total_ms, cost, saved in rows:
             db.add(ComparisonRecord(
@@ -59,44 +56,26 @@ def test_summary_uses_only_success_durations_and_filters_beijing_date() -> None:
             db, date_from=date(2038, 1, 15), date_to=date(2038, 1, 15)
         )
 
-        assert summary["started"] == 7
-        assert summary["completed"] == 5
-        assert summary["success"] == 3
-        assert summary["success_rate"] == pytest.approx(3 / 6)
+        assert summary["started"] == 4
+        assert summary["completed"] == 3
+        assert summary["success"] == 2
+        assert summary["success_rate"] == pytest.approx(2 / 3)
         assert summary["avg_token_cost"] == pytest.approx(2.5)
-        assert summary["lower_price_rate"] == pytest.approx(1 / 3)
-        assert summary["avg_duration_ms"] == 3000
-        assert summary["p5_duration_ms"] == 1200
-        assert summary["p50_duration_ms"] == 3000
-        assert summary["p95_duration_ms"] == 4800
-        assert summary["p99_duration_ms"] == 4960
+        assert summary["lower_price_rate"] == 0.5
+        assert summary["avg_duration_ms"] == 2000
+        assert summary["p5_duration_ms"] == 1100
+        assert summary["p50_duration_ms"] == 2000
+        assert summary["p95_duration_ms"] == 2900
+        assert summary["p99_duration_ms"] == 2980
         assert summary["cancelled"] == 1
-        assert summary["cancelled_rate"] == pytest.approx(1 / 7)
+        assert summary["cancelled_rate"] == 0.25
         assert summary["cancelled_p50_ms"] == 5000
 
         items, _next_cursor, total = queries.list_comparison_records(
             db, date_from=date(2038, 1, 15), date_to=date(2038, 1, 15), limit=20
         )
-        assert total == 7
+        assert total == 4
         assert {item.trace_id for item in items} == {row[0] for row in rows}
-
-        success_items, _next_cursor, success_total = queries.list_comparison_records(
-            db, status="success", date_from=date(2038, 1, 15), date_to=date(2038, 1, 15)
-        )
-        assert success_total == 3
-        assert {item.status for item in success_items} == {"success", "below_minimum"}
-
-        failed_items, _next_cursor, failed_total = queries.list_comparison_records(
-            db, status="failed", date_from=date(2038, 1, 15), date_to=date(2038, 1, 15)
-        )
-        assert failed_total == 2
-        assert {item.status for item in failed_items} == {"failed", "store_closed"}
-
-        running_items, _next_cursor, running_total = queries.list_comparison_records(
-            db, status="running", date_from=date(2038, 1, 15), date_to=date(2038, 1, 15)
-        )
-        assert running_total == 1
-        assert running_items[0].status == "running"
     finally:
         db.rollback()
         db.close()

tests/test_comparison_below_minimum_migration.py

@@ -1,95 +0,0 @@
-"""Regression coverage for comparison terminal-status normalization."""
-from __future__ import annotations
-
-import importlib.util
-from pathlib import Path
-
-import sqlalchemy as sa
-from alembic.migration import MigrationContext
-from alembic.operations import Operations
-
-
-def _migration_module():
-    path = (
-        Path(__file__).parents[1]
-        / "alembic"
-        / "versions"
-        / "comparison_below_minimum_as_success.py"
-    )
-    spec = importlib.util.spec_from_file_location(path.stem, path)
-    assert spec is not None and spec.loader is not None
-    module = importlib.util.module_from_spec(spec)
-    spec.loader.exec_module(module)
-    return module
-
-
-def test_business_status_migration_upgrade_and_downgrade() -> None:
-    engine = sa.create_engine("sqlite:///:memory:")
-    metadata = sa.MetaData()
-    records = sa.Table(
-        "comparison_record",
-        metadata,
-        sa.Column("id", sa.Integer, primary_key=True),
-        sa.Column("status", sa.String(16), nullable=False),
-        sa.Column("fail_reason", sa.String(256)),
-        sa.Column("raw_payload", sa.JSON, nullable=False),
-    )
-    metadata.create_all(engine)
-    module = _migration_module()
-
-    with engine.begin() as connection:
-        connection.execute(
-            records.insert(),
-            [
-                {
-                    "status": "below_minimum",
-                    "fail_reason": "旧失败原因",
-                    "raw_payload": {"record_status": "below_minimum"},
-                },
-                {
-                    "status": "success",
-                    "fail_reason": None,
-                    "raw_payload": {"record_status": "success"},
-                },
-                {
-                    "status": "failed",
-                    "fail_reason": "技术异常",
-                    "raw_payload": {"status": "failed"},
-                },
-                {
-                    "status": "store_closed",
-                    "fail_reason": "店铺打烊",
-                    "raw_payload": {"record_status": "store_closed"},
-                },
-                {
-                    "status": "items_not_found",
-                    "fail_reason": "商品未找到",
-                    "raw_payload": {"status": "items_not_found"},
-                },
-            ],
-        )
-        module.op = Operations(MigrationContext.configure(connection))
-
-        module.upgrade()
-        upgraded = connection.execute(
-            sa.select(records.c.status, records.c.fail_reason).order_by(records.c.id)
-        ).all()
-        assert upgraded == [
-            ("success", None),
-            ("success", None),
-            ("failed", "技术异常"),
-            ("failed", "店铺打烊"),
-            ("failed", "商品未找到"),
-        ]
-
-        module.downgrade()
-        downgraded = connection.execute(
-            sa.select(records.c.status, records.c.fail_reason).order_by(records.c.id)
-        ).all()
-        assert downgraded == [
-            ("below_minimum", None),
-            ("success", None),
-            ("failed", "技术异常"),
-            ("store_closed", "店铺打烊"),
-            ("items_not_found", "商品未找到"),
-        ]

tests/test_coupon_point_score.py

@@ -10,6 +10,7 @@ from app.admin.repositories.coupon_data import (
     _point_scores_by_trace,
     coupon_data_report,
     coupon_point_details,
+    coupon_user_records,
 )
 from app.admin.security import create_admin_token
 from app.db.session import SessionLocal
@@ -38,6 +39,7 @@ def test_point_scores_by_trace() -> None:
         stats = _point_scores_by_trace(db, [trace])[trace]
         assert stats["succeeded"] == 2
         assert stats["tried"] == 3
+        assert stats["events"] == 4
         details = coupon_point_details(db, trace_id=trace)
         assert [item["status"] for item in details] == [
             "success", "already_claimed", "failed", "skipped"
@@ -48,8 +50,8 @@ def test_point_scores_by_trace() -> None:
         db.close()
 
 
-def test_skipped_detail_does_not_create_a_score() -> None:
-    """仅有 skipped 时按需明细仍可查到，但列表没有虚假的 0/0 分数。"""
+def test_skipped_detail_is_distinguished_from_no_events() -> None:
+    """仅有 skipped 时分数仍为0/0，但保留事件数供前端开放明细。"""
     db = SessionLocal()
     trace = "point-score-skipped"
     try:
@@ -63,7 +65,7 @@ def test_skipped_detail_does_not_create_a_score() -> None:
         db.flush()
 
         scores = _point_scores_by_trace(db, [trace, "missing-trace"])
-        assert trace not in scores
+        assert scores[trace] == {"succeeded": 0, "tried": 0, "events": 1}
         assert "missing-trace" not in scores
         assert coupon_point_details(db, trace_id=trace)[0]["status"] == "skipped"
     finally:
@@ -114,6 +116,112 @@ def test_coupon_data_report_returns_scores_without_embedding_details() -> None:
         db.close()
 
 
+def test_coupon_data_report_marks_abandoned_without_point_results() -> None:
+    """中途退出且没有逐券终态时返回0/0，其他状态缺埋点仍保持为空。"""
+    db = SessionLocal()
+    report_date = date(2020, 1, 6)
+    user_id = 910006
+    try:
+        db.add_all([
+            CouponSession(
+                trace_id="point-score-abandoned-without-result",
+                device_id="score-abandoned-device",
+                user_id=user_id,
+                status="abandoned",
+                app_env="prod",
+                platforms=["meituan-waimai"],
+                started_at=datetime(2020, 1, 6, tzinfo=UTC),
+                started_date=report_date,
+            ),
+            CouponSession(
+                trace_id="point-score-completed-without-result",
+                device_id="score-completed-device",
+                user_id=user_id,
+                status="completed",
+                app_env="prod",
+                platforms=["meituan-waimai"],
+                started_at=datetime(2020, 1, 6, 1, tzinfo=UTC),
+                started_date=report_date,
+            ),
+            CouponSession(
+                trace_id="point-score-abandoned-with-result",
+                device_id="score-abandoned-result-device",
+                user_id=user_id,
+                status="abandoned",
+                app_env="prod",
+                platforms=["meituan-waimai"],
+                started_at=datetime(2020, 1, 6, 2, tzinfo=UTC),
+                started_date=report_date,
+            ),
+            CouponSession(
+                trace_id="point-score-abandoned-skipped-only",
+                device_id="score-abandoned-skipped-device",
+                user_id=user_id,
+                status="abandoned",
+                app_env="prod",
+                platforms=["meituan-waimai"],
+                started_at=datetime(2020, 1, 6, 3, tzinfo=UTC),
+                started_date=report_date,
+            ),
+        ])
+        db.add_all([
+            CouponClaimEvent(
+                trace_id="point-score-abandoned-with-result",
+                device_id="score-abandoned-result-device",
+                coupon_id=f"mt-abandoned-{status}",
+                claim_date=report_date,
+                status=status,
+            )
+            for status in ("success", "failed")
+        ])
+        db.add(CouponClaimEvent(
+            trace_id="point-score-abandoned-skipped-only",
+            device_id="score-abandoned-skipped-device",
+            coupon_id="mt-abandoned-skipped",
+            claim_date=report_date,
+            status="skipped",
+        ))
+        db.flush()
+
+        report = coupon_data_report(
+            db,
+            date_from=report_date.isoformat(),
+            date_to=report_date.isoformat(),
+            app_env="prod",
+        )
+        rows = {item["trace_id"]: item for item in report["items"]}
+        abandoned = rows["point-score-abandoned-without-result"]
+        assert abandoned["point_success_count"] == 0
+        assert abandoned["point_total_count"] == 0
+        assert abandoned["point_event_count"] == 0
+
+        abandoned_with_result = rows["point-score-abandoned-with-result"]
+        assert abandoned_with_result["point_success_count"] == 1
+        assert abandoned_with_result["point_total_count"] == 2
+        assert abandoned_with_result["point_event_count"] == 2
+
+        abandoned_skipped = rows["point-score-abandoned-skipped-only"]
+        assert abandoned_skipped["point_success_count"] == 0
+        assert abandoned_skipped["point_total_count"] == 0
+        assert abandoned_skipped["point_event_count"] == 1
+
+        completed = rows["point-score-completed-without-result"]
+        assert completed["point_success_count"] is None
+        assert completed["point_total_count"] is None
+
+        user_rows = {
+            item["trace_id"]: item
+            for item in coupon_user_records(db, user_id=user_id)["items"]
+        }
+        assert user_rows["point-score-abandoned-without-result"]["point_total_count"] == 0
+        assert user_rows["point-score-abandoned-with-result"]["point_total_count"] == 2
+        assert user_rows["point-score-abandoned-skipped-only"]["point_event_count"] == 1
+        assert user_rows["point-score-completed-without-result"]["point_total_count"] is None
+    finally:
+        db.rollback()
+        db.close()
+
+
 def test_coupon_point_details_endpoint() -> None:
     """前端点击使用的接口按约定返回 trace_id 和逐券 items。"""
     db = SessionLocal()