@sophiacyl/dsh-plugin-waves 0.0.0-stage → 0.1.0-beta.1

package/service/rover/webconsole/mount_app.py

@@ -0,0 +1,26 @@
+"""后台管理注册桩。吃下插件的声明, 不提供 Web 界面。"""
+
+from dataclasses import dataclass
+from typing import Any, Type
+
+
+@dataclass
+class PageSchema:
+    label: str
+    icon: str = ""
+
+
+class AdminModel:
+    model: Any = None
+    pk_name: str = "id"
+    page_schema: Any = None
+
+
+class Site:
+    def register_admin(self, *admin_cls: Type[AdminModel], _ADD: bool = False) -> Type[AdminModel]:
+        return admin_cls[0]
+
+
+site = Site()
+
+__all__ = ["PageSchema", "AdminModel", "Site", "site"]

package/service/tools/migrate.py

@@ -0,0 +1,351 @@
+#!/usr/bin/env python3
+"""从既有部署迁入配置、凭据与素材。默认只做缺失项, --force 覆盖。"""
+
+import argparse
+import json
+import shutil
+import sqlite3
+import sys
+from pathlib import Path
+from typing import Optional
+
+DST_DATA = Path(__file__).resolve().parents[1] / "data"
+
+# 配置模型类名去掉了 Gs 前缀, 存量 json 的 type 标签要跟着换
+TYPE_MAP = {
+    f"Gs{n}Config": f"{n}Config"
+    for n in ("Str", "Bool", "Dict", "ListStr", "List", "Int", "Float", "Image",
+              "TimeR", "FileUpload", "FilesUpload", "Date", "TimeRange", "Color",
+              "RepeatGroup", "Time")
+}
+TYPE_MAP["GsDivider"] = "Divider"
+
+PLUGINS = ["XutheringWavesUID", "RoverSign", "ScoreEcho"]
+
+SYMLINK_DIRS = ["guide_new", "custom_role_pile", "custom_mr_bg", "custom_orb", "custom_mr_role_pile"]
+COPY_DIRS = ["alias", "show", "resource"]
+COPY_FILES = ["ann_data.json", "guide_config.json", "gacha_config.json"]
+
+# 订阅表不迁: 那是群聊场景的推送去向, 本地只有一个使用者, 迁过来只会留下外部群号
+DB_TABLES = [
+    "wavesbind", "wavesuser", "WavesUserSdk", "WavesGachaCloud",
+    "WavesLangSettings", "WavesStaminaRecord", "RoverSign", "ScoreEcho",
+    "ScoreLangSettings",
+]
+
+
+def _rewrite_paths(item: dict, src_root: Path, dst_root: Path) -> int:
+    """配置里存的是绝对路径, 迁过来要落到本地数据目录的同名位置。"""
+    changed = 0
+    prefix = str(src_root)
+    for field in ("data", "upload_to"):
+        value = item.get(field)
+        if not isinstance(value, str) or not value:
+            continue
+        if value.startswith(prefix):
+            item[field] = str(dst_root) + value[len(prefix) :]
+        elif "/XutheringWavesUID/" in value.replace("\\", "/"):
+            # 来源部署的目录结构不同时, 按插件目录之后的相对部分归位
+            tail = value.replace("\\", "/").split("/XutheringWavesUID/", 1)[1]
+            item[field] = str(dst_root / "XutheringWavesUID" / tail)
+        else:
+            continue
+        changed += 1
+    return changed
+
+
+def convert_config(src: Path, dst: Path, force: bool, roots: Optional[tuple] = None) -> None:
+    if not src.is_file():
+        print(f"跳过 (源缺失): {src}")
+        return
+    if dst.exists() and not force:
+        print(f"跳过 (已存在): {dst}")
+        return
+    data = json.loads(src.read_text("utf-8"))
+    paths = 0
+    for item in data.values():
+        if not isinstance(item, dict):
+            continue
+        if item.get("type") in TYPE_MAP:
+            item["type"] = TYPE_MAP[item["type"]]
+        if roots:
+            paths += _rewrite_paths(item, *roots)
+    if paths:
+        print(f"路径改写: {paths} 处指向本地数据目录")
+    dst.parent.mkdir(parents=True, exist_ok=True)
+    dst.write_text(json.dumps(data, ensure_ascii=False, indent=2), "utf-8")
+    secrets = [k for k in ("WavesToken", "LocalProxyUrl", "xwtoken") if k in data]
+    print(f"配置: {dst.name} 项数={len(data)} 含凭据={secrets}")
+
+
+def _prune_users(conn: sqlite3.Connection, masters: list) -> int:
+    """只留主人自己的行, 其余用户数据不进本地库。"""
+    if not masters:
+        return 0
+    placeholders = ",".join("?" for _ in masters)
+    removed = 0
+    for table in ("wavesbind", "wavesuser", "WavesUserSdk", "WavesGachaCloud",
+                  "WavesLangSettings", "WavesStaminaRecord", "ScoreEcho",
+                  "ScoreLangSettings"):
+        try:
+            cur = conn.execute(
+                f'DELETE FROM "{table}" WHERE user_id IS NOT NULL '
+                f"AND user_id NOT IN ({placeholders})",
+                masters,
+            )
+            removed += cur.rowcount or 0
+        except sqlite3.OperationalError:
+            continue
+    conn.commit()
+    return removed
+
+
+def find_source_db(src: Path) -> Optional[Path]:
+    """来源库文件名随部署而异, 按是否含绑定表识别。"""
+    for path in sorted(src.glob("*.db")):
+        try:
+            conn = sqlite3.connect(f"file:{path}?mode=ro&immutable=1", uri=True)
+        except sqlite3.Error:
+            continue
+        try:
+            names = {
+                str(row[0]).lower()
+                for row in conn.execute("SELECT name FROM sqlite_master WHERE type='table'")
+            }
+        except sqlite3.Error:
+            names = set()
+        finally:
+            conn.close()
+        if "wavesbind" in names:
+            return path
+    return None
+
+
+def snapshot_db(src: Optional[Path], dst: Path, force: bool, masters: list) -> None:
+    if src is None or not src.is_file():
+        print(f"跳过 (源库缺失): {src}")
+        return
+    if dst.exists() and not force:
+        print(f"跳过 (已存在): {dst}")
+        return
+    tmp = dst.with_suffix(".snapshot")
+    for path in (tmp, Path(str(tmp) + "-wal"), Path(str(tmp) + "-shm")):
+        path.unlink(missing_ok=True)
+    # 只读 + 快照, 不与在写进程共享连接
+    source = sqlite3.connect(f"file:{src}?mode=ro&immutable=1", uri=True)
+    target = sqlite3.connect(tmp)
+    try:
+        source.backup(target)
+    finally:
+        target.close()
+        source.close()
+
+    keep = sqlite3.connect(tmp)
+    try:
+        existing = {r[0] for r in keep.execute("SELECT name FROM sqlite_master WHERE type='table'")}
+        dropped = 0
+        for table in existing:
+            if table not in DB_TABLES and not table.startswith("sqlite_"):
+                keep.execute(f'DROP TABLE IF EXISTS "{table}"')
+                dropped += 1
+        keep.commit()
+        pruned = _prune_users(keep, masters)
+        rows = {t: keep.execute(f'SELECT COUNT(*) FROM "{t}"').fetchone()[0]
+                for t in DB_TABLES if t in existing}
+        keep.execute("VACUUM")
+        keep.commit()
+        print(f"数据库: 丢弃他人行={pruned}")
+    finally:
+        keep.close()
+    tmp.replace(dst)
+    print(f"数据库: {dst.name} 保留表={rows} 丢弃无关表={dropped}")
+
+
+def rewrite_bot_id(db: Path, bot_id: str = "dsh") -> None:
+    """来源库按原平台标识分区, 本地只有一个会话来源, 统一改写。"""
+    if not db.is_file():
+        return
+    conn = sqlite3.connect(db)
+    try:
+        changed = 0
+        for table in DB_TABLES:
+            for column in ("bot_id", "bot_self_id", "WS_BOT_ID"):
+                try:
+                    cur = conn.execute(
+                        f'UPDATE "{table}" SET "{column}"=? WHERE "{column}" IS NOT NULL '
+                        f'AND "{column}"!=?',
+                        (bot_id, bot_id),
+                    )
+                    changed += cur.rowcount or 0
+                except sqlite3.OperationalError:
+                    continue
+        conn.commit()
+        print(f"会话来源: {changed} 处改写为 {bot_id}")
+    finally:
+        conn.close()
+
+
+def rewrite_user_id(db: Path, user_id: str) -> None:
+    """来源库按聊天平台的用户 id 分区, 本地只有一个使用者, 统一归到它名下。"""
+    if not db.is_file() or not user_id:
+        return
+    conn = sqlite3.connect(db)
+    try:
+        changed = 0
+        for table in DB_TABLES:
+            try:
+                cur = conn.execute(
+                    f'UPDATE "{table}" SET user_id=? WHERE user_id IS NOT NULL AND user_id!=?',
+                    (user_id, user_id),
+                )
+                changed += cur.rowcount or 0
+            except sqlite3.OperationalError:
+                continue
+        conn.commit()
+        print(f"归属: {changed} 处改写为 {user_id}")
+    finally:
+        conn.close()
+
+
+def bind_group(db: Path, group: str = "dsh") -> None:
+    """把绑定记录挂到本地会话所属的组, 群维度的排行才有数据。"""
+    if not db.is_file():
+        return
+    conn = sqlite3.connect(db)
+    try:
+        rows = conn.execute("SELECT id, group_id FROM wavesbind").fetchall()
+        changed = 0
+        for row_id, group_id in rows:
+            parts = [g for g in (group_id or "").split("_") if g]
+            if group not in parts:
+                parts.append(group)
+                conn.execute("UPDATE wavesbind SET group_id=? WHERE id=?", ("_".join(parts), row_id))
+                changed += 1
+        conn.commit()
+        print(f"绑定: {changed}/{len(rows)} 行加入 {group}")
+    finally:
+        conn.close()
+
+
+def master_uids(db: Path) -> list:
+    if not db.is_file():
+        return []
+    conn = sqlite3.connect(db)
+    try:
+        uids = set()
+        for (cell,) in conn.execute("SELECT uid FROM wavesbind WHERE uid IS NOT NULL"):
+            uids.update(u for u in str(cell).split("_") if u)
+        for (cell,) in conn.execute("SELECT uid FROM wavesuser WHERE uid IS NOT NULL"):
+            if cell:
+                uids.add(str(cell))
+        return sorted(uids)
+    finally:
+        conn.close()
+
+
+def copy_players(src: Path, dst: Path, uids: list, force: bool) -> None:
+    if not src.is_dir() or not uids:
+        print("跳过 players (无绑定 uid)")
+        return
+    dst.mkdir(parents=True, exist_ok=True)
+    copied = 0
+    for uid in uids:
+        one_src, one_dst = src / uid, dst / uid
+        if not one_src.is_dir():
+            continue
+        if one_dst.exists():
+            if not force:
+                continue
+            shutil.rmtree(one_dst)
+        shutil.copytree(one_src, one_dst, symlinks=True)
+        copied += 1
+    print(f"玩家数据: {copied}/{len(uids)} 个 uid")
+
+
+def link_or_copy(src: Path, dst: Path, mode: str, force: bool) -> None:
+    if not src.exists():
+        print(f"跳过 (源缺失): {src.name}")
+        return
+    if dst.exists() or dst.is_symlink():
+        if not force:
+            print(f"跳过 (已存在): {dst.name}")
+            return
+        if dst.is_symlink() or dst.is_file():
+            dst.unlink()
+        else:
+            shutil.rmtree(dst)
+    dst.parent.mkdir(parents=True, exist_ok=True)
+    if mode == "symlink":
+        try:
+            dst.symlink_to(src)
+            print(f"链接: {dst.name} -> {src}")
+            return
+        except OSError:
+            # Windows 未开开发者模式时创建符号链接会失败, 退回复制
+            print(f"链接失败, 改为复制: {dst.name}")
+    if src.is_dir():
+        shutil.copytree(src, dst, symlinks=True)
+        print(f"复制: {dst.name}/")
+    else:
+        shutil.copy2(src, dst)
+        print(f"复制: {dst.name}")
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--src", type=Path, help="来源部署的 data 目录, 不给则跳过迁入")
+    parser.add_argument("--src-db", type=Path, help="来源数据库, 默认在 --src 下自动识别")
+    parser.add_argument("--force", action="store_true")
+    parser.add_argument("--skip-players", action="store_true")
+    parser.add_argument("--masters", nargs="*", help="要保留的用户 id, 默认取 data/config.json")
+    args = parser.parse_args()
+    src = args.src
+    if src is None:
+        print("未指定 --src, 跳过迁入 (全新部署无需迁移)")
+        return
+    if not src.is_dir():
+        sys.exit(f"源数据目录不存在: {src}")
+
+    DST_DATA.mkdir(parents=True, exist_ok=True)
+
+    for plugin in PLUGINS:
+        convert_config(
+            src / plugin / "config.json",
+            DST_DATA / plugin / "config.json",
+            args.force,
+            (src.resolve(), DST_DATA.resolve()),
+        )
+
+    masters = args.masters or json.loads(
+        (DST_DATA / "config.json").read_text("utf-8")
+    ).get("masters", [])
+    snapshot_db(
+        args.src_db or find_source_db(src),
+        DST_DATA / "WavesData.db",
+        args.force,
+        masters,
+    )
+    rewrite_bot_id(DST_DATA / "WavesData.db")
+    rewrite_user_id(DST_DATA / "WavesData.db", masters[0] if masters else "local")
+    bind_group(DST_DATA / "WavesData.db")
+
+    ww_src, ww_dst = src / "XutheringWavesUID", DST_DATA / "XutheringWavesUID"
+    for name in COPY_DIRS:
+        link_or_copy(ww_src / name, ww_dst / name, "copy", args.force)
+    for name in SYMLINK_DIRS:
+        link_or_copy(ww_src / name, ww_dst / name, "symlink", args.force)
+    for name in COPY_FILES:
+        link_or_copy(ww_src / name, ww_dst / name, "copy", args.force)
+    if not args.skip_players:
+        copy_players(
+            ww_src / "players",
+            ww_dst / "players",
+            master_uids(DST_DATA / "WavesData.db"),
+            args.force,
+        )
+
+    print("\n完成。若首次运行, 建议接着执行: .venv/bin/python tools/smoke.py ww查询")
+
+
+if __name__ == "__main__":
+    main()

package/service/tools/regression.py

@@ -0,0 +1,87 @@
+#!/usr/bin/env python3
+"""跑一批命令, 汇总命中与回复形态。默认打到本机服务。
+
+只跑只读命令: 删除/清理/压缩/上传类会改动素材与数据库, 一律拦下。
+"""
+
+import argparse
+import json
+import sys
+import urllib.error
+import urllib.request
+
+COMMANDS = [
+    "ww帮助", "ww查询", "ww体力", "ww探索", "ww库洛币", "ww积分", "ww日历", "ww签到日历",
+    "ww深塔", "ww海墟", "ww矩阵", "ww全息",
+    "ww练度", "ww声骸", "ww星声",
+    "ww椿面板", "ww椿伤害", "ww椿权重", "ww椿优化", "ww椿总排行", "ww椿声骸总排行",
+    "ww椿攻略", "ww椿图鉴", "ww椿养成", "ww武器列表", "ww套装列表",
+    "ww练度总排行", "ww无尽总排行", "ww矩阵总排行",
+    "ww抽卡记录", "ww抽卡帮助", "ww兑换码", "ww卡池倒计时",
+    "ww持有率", "ww深塔出场率", "ww矩阵配队",
+    "ww签到", "ww签到帮助", "ww分析帮助", "ww综合评分说明",
+    "ww查看特征码", "ww更新记录", "ww别名列表", "ww公告",
+]
+
+
+# 会改动素材/数据库的命令, 不允许出现在自检里
+DESTRUCTIVE = ("删除", "清理", "清除", "压缩", "上传", "重新计算", "强制下载", "导入")
+
+
+def guard(commands: list) -> list:
+    blocked = [c for c in commands if any(word in c for word in DESTRUCTIVE)]
+    if blocked:
+        sys.exit("这些命令有副作用, 不能用于自检: " + " ".join(blocked))
+    return commands
+
+
+def run(base: str, command: str, timeout: int) -> dict:
+    body = json.dumps({"command": command}).encode()
+    req = urllib.request.Request(
+        f"{base}/api/chat", data=body, headers={"content-type": "application/json"}
+    )
+    try:
+        with urllib.request.urlopen(req, timeout=timeout) as resp:
+            return json.loads(resp.read())
+    except urllib.error.HTTPError as e:
+        return {"ok": False, "summary": f"HTTP {e.code}", "segments": [], "matched": []}
+    except Exception as e:
+        return {"ok": False, "summary": f"{type(e).__name__}: {e}", "segments": [], "matched": []}
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--base", default="http://127.0.0.1:9777")
+    parser.add_argument("--timeout", type=int, default=180)
+    parser.add_argument("commands", nargs="*")
+    args = parser.parse_args()
+
+    commands = guard(args.commands or COMMANDS)
+    stats = {"html": 0, "image": 0, "text": 0, "node": 0, "other": 0}
+    unmatched, failed = [], []
+
+    for command in commands:
+        data = run(args.base, command, args.timeout)
+        segs = data.get("segments") or []
+        kinds = [s.get("type") for s in segs]
+        for kind in kinds:
+            stats[kind if kind in stats else "other"] += 1
+        summary = (data.get("summary") or "").replace("\n", " ")[:70]
+        flag = "√" if data.get("ok") else "×"
+        if not data.get("ok"):
+            unmatched.append(command)
+        elif not kinds:
+            failed.append(command)
+        print(f"{flag} {command:<16} {','.join(kinds) or '-':<22} {summary}")
+
+    print(f"\n段统计: {stats}")
+    if unmatched:
+        print(f"未命中/失败 ({len(unmatched)}): {' '.join(unmatched)}")
+    if failed:
+        print(f"命中但无回复 ({len(failed)}): {' '.join(failed)}")
+    print(f"总计 {len(commands)} 条, 正常 {len(commands) - len(unmatched) - len(failed)} 条")
+    sys.exit(1 if unmatched or failed else 0)
+
+
+if __name__ == "__main__":
+    main()

package/service/tools/smoke.py

@@ -0,0 +1,63 @@
+#!/usr/bin/env python3
+"""不起 HTTP 服务, 直接跑一遍启动流程与若干命令。"""
+
+import argparse
+import asyncio
+import sys
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).resolve().parents[1]))
+
+
+def local_user() -> str:
+    """本机使用者标识, 从配置里取, 不写死。"""
+    from rover.config import core_config
+
+    masters = core_config.get_config("masters") or []
+    return str(masters[0]) if masters else "local"
+
+
+async def boot():
+    from rover.app_life import load_plugins
+    from rover.server import core_start_before_execute
+    from rover.utils.database.base_models import init_database
+    from rover.utils.database.startup import init_all
+
+    await init_database()
+    load_plugins()
+    await init_all()
+    await core_start_before_execute()
+
+
+async def run_commands(commands):
+    from rover.api_render import render_segments, summarize
+    from rover.handler import build_message, dispatch, normalize_command
+    from rover.sv import SL
+
+    print(f"\n注册: 插件={list(SL.plugins.keys())} SV={len(SL.lst)} "
+          f"触发器={sum(len(t) for sv in SL.lst.values() for t in sv.TL.values())}")
+
+    for raw in commands:
+        command = normalize_command(raw)
+        result = await dispatch(build_message(command, user_id=local_user()))
+        rendered = await render_segments(result["segments"])
+        kinds = [s["type"] for s in rendered]
+        print(f"\n[{command}] 命中={result['matched']} 段={kinds} 用时={result['elapsed']:.2f}s")
+        text = summarize(rendered)
+        if text:
+            print("  " + text.replace("\n", "\n  ")[:600])
+
+
+async def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("commands", nargs="*", default=[])
+    args = parser.parse_args()
+    await boot()
+    if args.commands:
+        await run_commands(args.commands)
+    else:
+        await run_commands([])
+
+
+if __name__ == "__main__":
+    asyncio.run(main())