yunke/yunke-server/app/api/schemas.py

"""Pydantic v2 response models for the yunke data-mirror API."""

from __future__ import annotations

import datetime as dt
from typing import Any, Generic, Literal, TypeVar

from pydantic import BaseModel, ConfigDict

T = TypeVar("T")


class CustomerOut(BaseModel):
    """Public Customer view — a subset of :class:`app.db.models.Customer`.

    The structured columns power the list table; the verbatim upstream record
    is exposed as ``raw_json`` so the frontend can render the full ~60-field
    云客 客户详情 (all-fields drawer) without a second endpoint. The model's
    ``raw_json`` column is picked up automatically via ``from_attributes``.
    """

    model_config = ConfigDict(from_attributes=True)

    friend_id: str
    channel: str
    owner_account_id: str | None = None
    friend_nickname: str | None = None
    friend_remark: str | None = None
    real_name: str | None = None
    phone: str | None = None
    region: str | None = None
    sale_name: str | None = None
    dept_name: str | None = None
    head_image_url: str | None = None
    last_chat_time: int | None = None
    add_time: int | None = None
    message_count: int | None = None
    version: str | None = None
    # Verbatim upstream friend record (~60 camelCase fields). Surfaced so the
    # all-fields drawer can render every column the 云客 source page shows
    # (friendWxAlias/addTime/createTime/fromType/addType/...) without the
    # backend having to enumerate each one as a structured column.
    raw_json: dict[str, Any] | None = None


class Page(BaseModel, Generic[T]):
    """Generic pagination envelope returned by list endpoints."""

    items: list[T]
    total: int
    page: int
    page_size: int


# ---------------------------------------------------------------------------
# Organization & people
# ---------------------------------------------------------------------------


class DepartmentOut(BaseModel):
    """云客 department node — frontend builds the tree from ``parent_id``/``path``."""

    model_config = ConfigDict(from_attributes=True)

    dept_id: str
    parent_id: str | None = None
    name: str | None = None
    path: str | None = None


class StaffOut(BaseModel):
    """Staff member (employee) row.

    The structured columns power the list table; the verbatim upstream record
    is exposed as ``raw_json`` so the frontend can render the all-fields drawer
    without a second endpoint. Picked up automatically via ``from_attributes``.
    """

    model_config = ConfigDict(from_attributes=True)

    user_id: str
    name: str | None = None
    phone: str | None = None
    sale_phone: str | None = None
    dept_id: str | None = None
    is_on_job: bool | None = None
    raw_json: dict[str, Any] | None = None


class AccountOut(BaseModel):
    """Unified staff messaging account (WeChat or WhatsApp).

    The two source tables (:class:`WechatAccount`, :class:`WhatsappAccount`)
    have different PK / column names; this view projects both onto a single
    shape keyed by ``channel`` + ``account_id`` so the frontend can render one
    list. Built explicitly in the route handler (not via ``from_attributes``)
    because no single ORM row carries this schema. ``raw_json`` carries the
    verbatim upstream record (wechatId/alias/sendNum/receiveNum/dbSize/...) so
    the all-fields drawer can render every column the 云客 微信号管理 page shows.
    """

    channel: str
    account_id: str
    nickname: str | None = None
    user_id: str | None = None
    user_name: str | None = None
    phone: str | None = None
    contacts_num: int | None = None
    group_num: int | None = None
    is_on_job: bool | None = None
    raw_json: dict[str, Any] | None = None


# ---------------------------------------------------------------------------
# Communication entities
# ---------------------------------------------------------------------------


class MessageOut(BaseModel):
    """Chat timeline row — a single message (text/media/voice/...).

    The structured columns power the bubble rendering; the verbatim upstream
    record is exposed as ``raw_json`` so the frontend can render the all-fields
    drawer (msgSvrId/keyId, file/fileTh/fileHd, md5/md5Th/md5Hd, paymsgid,
    referMsg, imgPath, ...) without a second endpoint. Picked up automatically
    via ``from_attributes``.
    """

    model_config = ConfigDict(from_attributes=True)

    msg_id: str
    channel: str
    owner_account_id: str | None = None
    talker_id: str | None = None
    room_id: str | None = None
    is_send: bool | None = None
    type: str | None = None
    type_raw: int | None = None
    text: str | None = None
    file_url: str | None = None
    # Image thumbnails / originals + voice duration / transcription so the
    # timeline can render media without a second round-trip. All optional — the
    # underlying columns (see Message model) are nullable.
    file_th_url: str | None = None
    file_hd_url: str | None = None
    duration: int | None = None
    transcription_text: str | None = None
    md5: str | None = None
    sender_name: str | None = None
    sender_head_url: str | None = None
    timestamp: int | None = None
    # sha256 media_hash of the localized media blob (NULL until the
    # message_images sync pass downloads it). When present, the frontend
    # renders via ``/api/media/{media_hash}`` (local-first); when absent, it
    # falls back to the OSS ``file_url`` / ``file_th_url`` / ``file_hd_url``.
    media_hash: str | None = None
    raw_json: dict[str, Any] | None = None


class GroupOut(BaseModel):
    """Chat group summary row.

    The structured columns power the list table; the verbatim upstream record
    is exposed as ``raw_json`` so the frontend can render the all-fields drawer
    (chatRoomId/chatRoomName, memberIds, headImageUrl, wechatId/wechatNickName,
    userName/userPhone, messageCount/ownMessageCount, lossMembers/addMembers,
    lastChatTime/updateTime, ...) without a second endpoint. Picked up
    automatically via ``from_attributes``.
    """

    model_config = ConfigDict(from_attributes=True)

    room_id: str
    channel: str
    room_name: str | None = None
    owner_account_id: str | None = None
    owner_account_nickname: str | None = None
    member_count: int | None = None
    message_count: int | None = None
    last_chat_time: int | None = None
    raw_json: dict[str, Any] | None = None


class GroupMemberOut(BaseModel):
    """Member of a chat group (composite PK: channel + room_id + member_id)."""

    model_config = ConfigDict(from_attributes=True)

    channel: str
    room_id: str
    member_id: str
    nickname: str | None = None
    display_name: str | None = None
    avatar: str | None = None


class PhoneCallOut(BaseModel):
    """Phone / voice call record. ``channel`` is the call-type.

    The structured columns power the list table; the verbatim upstream record
    is exposed as ``raw_json`` so the frontend can render the all-fields drawer
    (id/startCallTime, callDuration/callSeconds, ringTime, callStatus,
    incomingCall, userId/userIdName, departmentId, callNumber,
    simPhone/simNumber/imei, recordFile, isUpload, downloadCount, aiAsr,
    qcStatus, planCustomerName, version + voice-call extras ossFileName/
    duration/startTime/endTime/remark) without a second endpoint. Picked up
    automatically via ``from_attributes``.
    """

    model_config = ConfigDict(from_attributes=True)

    call_id: str
    channel: str
    start_call_time: int | None = None
    call_duration: int | None = None
    call_status: int | None = None
    incoming_call: int | None = None
    user_id: str | None = None
    user_id_name: str | None = None
    call_number: str | None = None
    sim_phone: str | None = None
    # OSS URL of the call recording — surfaced so the Calls view can offer a
    # "播放录音" action when present.
    record_file: str | None = None
    # sha256 media_hash of the localized recording (NULL until downloaded).
    record_media_hash: str | None = None
    raw_json: dict[str, Any] | None = None


class SmsOut(BaseModel):
    """SMS log row.

    The structured columns power the list table; the verbatim upstream record
    is exposed as ``raw_json`` so the frontend can render the all-fields drawer
    (id/customerPhone, sms/smsSub, sender, direction, userId,
    senderDepartment, receiveName, createTime, simPhone/simNumber, source)
    without a second endpoint. Picked up automatically via ``from_attributes``.
    """

    model_config = ConfigDict(from_attributes=True)

    sms_id: str
    customer_phone: str | None = None
    content_preview: str | None = None
    # Full SMS body (the table stores it in ``content``); included so the SMS
    # view can show the whole message in a detail dialog without a second call.
    content: str | None = None
    sender: str | None = None
    direction: int | None = None
    user_id: str | None = None
    receive_name: str | None = None
    create_time: int | None = None
    sim_phone: str | None = None
    raw_json: dict[str, Any] | None = None


class VoiceClipOut(BaseModel):
    """长语音 (voice_clip) row — a long audio segment from ``/wechatweb/call/list``.

    The structured columns power the VoiceClipModal table (员工姓名 /
    企业串码 / 文件地址 / 录音开始时间 / 录音结束时间 / 录音时长); the verbatim
    upstream record is exposed as ``raw_json`` so the frontend can render any
    extra fields without a second endpoint. Picked up automatically via
    ``from_attributes``.
    """

    model_config = ConfigDict(from_attributes=True)

    voice_clip_id: int
    user_id: str
    channel: str
    user_real_name: str | None = None
    wx_id: str | None = None
    talker: str | None = None
    status: str | None = None
    is_send: bool | None = None
    company_code: str | None = None
    start_time: int | None = None
    end_time: int | None = None
    # OSS recording URL — surfaced so the modal can render an inline
    # ``<audio controls>`` player (see spec §6).
    oss_file_name: str | None = None
    # sha256 media_hash of the localized audio (NULL until downloaded). Playback
    # prefers /api/media/{this} over the OSS oss_file_name URL (local-first).
    media_hash: str | None = None
    mp3_duration: int | None = None
    duration_format: str | None = None
    # LOCAL-only 手动匹配 (spec §7): the customer.friend_id this clip has been
    # associated with, or None when unmatched. ``matched_friend_name`` is NOT a
    # stored column — it's resolved by the GET handler via a Customer lookup
    # (remark > nickname > real_name > friend_id) so the modal can render the
    # 「已匹配客户」 column without a second round-trip per row.
    matched_friend_id: str | None = None
    matched_friend_name: str | None = None
    raw_json: dict[str, Any] | None = None


# ---------------------------------------------------------------------------
# Sync monitor
# ---------------------------------------------------------------------------


class SyncStateOut(BaseModel):
    """Resumable sync cursor (one per entity/channel/scope)."""

    model_config = ConfigDict(from_attributes=True)

    state_id: int
    entity_type: str
    channel: str | None = None
    scope_id: str | None = None
    last_version: str | None = None
    last_timestamp: int | None = None
    last_synced_at: dt.datetime | None = None


class SyncRunOut(BaseModel):
    """Audit row for a single sync execution."""

    model_config = ConfigDict(from_attributes=True)

    run_id: int
    entity_type: str
    channel: str | None = None
    scope_id: str | None = None
    status: str
    started_at: dt.datetime
    finished_at: dt.datetime | None = None
    records_fetched: int
    records_upserted: int
    error: str | None = None


class SyncStatusOut(BaseModel):
    """Aggregate monitor payload — current cursors plus recent runs."""

    states: list[SyncStateOut]
    recent_runs: list[SyncRunOut]


# ---------------------------------------------------------------------------
# Staff chat stats (WeChat staff 5-number summary + customer detail)
# ---------------------------------------------------------------------------


class StaffChatCustomerOut(BaseModel):
    """单个客户在日期窗内的私聊聚合行（本地 message 表派生）。"""

    talker_id: str
    friend_name: str  # remark > nickname > real_name > talker_id
    send_count: int  # 员工发出（is_send True）
    receive_count: int  # 客户发出（is_send False）
    interaction_days: int  # 窗口内不同日期数
    last_is_send: int | None  # 1=员工最后发, 0=客户最后发, None=最后一条方向未知（不计入两率）
    last_time: int | None = None  # ms-epoch，最后一条时间


class StaffChatStatsSummary(BaseModel):
    """5 个汇总数字（从 customers 派生）。"""

    chat_count: int  # 沟通人数：窗口内 ≥1 条消息的去重客户数
    effective_count: int  # 有效沟通数：send≥1 且 receive≥1 的客户数
    follow_count: int  # 跟进客户数：interaction_days≥2 的客户数（近似口径）
    last_reply_count: int  # 员工最后回复数：last_is_send=1 的客户数
    no_reply_count: int  # 未回复客户数：last_is_send=0 的客户数


class StaffChatStatsOut(BaseModel):
    owner_account_id: str
    start: int  # ms-epoch
    end: int  # ms-epoch
    summary: StaffChatStatsSummary
    customers: list[StaffChatCustomerOut]


class SessionOut(BaseModel):
    """账号会话列表的一行(好友或群)。"""

    type: Literal["friend", "group"]
    talker_id: str | None = None
    room_id: str | None = None
    name: str | None = None
    avatar: str | None = None
    today_count: int
    last_chat_time: int | None = None


class SessionListOut(BaseModel):
    owner_account_id: str
    channel: str
    today_start: int
    sessions: list[SessionOut]