✨支持资源点查询

✨支持资源点查询

model/helpers.py

@@ -16,7 +16,7 @@ from service.cache import RedisCache
 
 USER_AGENT: str = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) " \
                   "Chrome/90.0.4430.72 Safari/537.36"
-headers: dict = {'User-Agent': USER_AGENT}
+REQUEST_HEADERS: dict = {'User-Agent': USER_AGENT}
 current_dir = os.getcwd()
 cache_dir = os.path.join(current_dir, "cache")
 if not os.path.exists(cache_dir):
@@ -56,7 +56,7 @@ async def url_to_file(url: str, prefix: str = "file://") -> str:
     temp_file_name = url_sha1 + extension
     file_dir = os.path.join(cache_dir, temp_file_name)
     if not os.path.exists(file_dir):
-        async with httpx.AsyncClient(headers=headers) as client:
+        async with httpx.AsyncClient(headers=REQUEST_HEADERS) as client:
             try:
                 data = await client.get(url)
             except UnsupportedProtocol as error:

plugins/map.py

@@ -0,0 +1,71 @@
+from os import sep
+
+from PIL import Image
+from telegram import Update
+from telegram.constants import ChatAction
+from telegram.ext import CommandHandler, MessageHandler, filters
+
+from logger import Log
+from manager import listener_plugins_class
+from plugins.base import BasePlugins, restricts
+from plugins.errorhandler import conversation_error_handler
+from service.map import MapHelper
+from utils.base import PaimonContext
+
+
+@listener_plugins_class()
+class Map(BasePlugins):
+
+    def __init__(self):
+        self.init_resource_map = False
+        self.map_helper = MapHelper()
+
+    @classmethod
+    def create_handlers(cls) -> list:
+        map_res = cls()
+        return [
+            CommandHandler("map", map_res.command_start, block=False),
+            MessageHandler(filters.Regex(r"^资源点查询(.*)"), map_res.command_start, block=True)
+        ]
+
+    async def init_point_list_and_map(self):
+        Log.info("正在初始化地图资源节点")
+        if not self.init_resource_map:
+            await self.map_helper.init_point_list_and_map()
+            self.init_resource_map = True
+
+    @conversation_error_handler
+    @restricts(restricts_time=20)
+    async def command_start(self, update: Update, context: PaimonContext):
+        message = update.message
+        args = context.args
+        user = update.effective_user
+        if not self.init_resource_map:
+            await self.init_point_list_and_map()
+        await message.reply_chat_action(ChatAction.TYPING)
+        if len(args) >= 1:
+            resource_name = args[0]
+        else:
+            Log.info(f"用户: {user.full_name} [{user.id}] 使用了 map 命令")
+            await message.reply_text("请输入要查找的资源，或私聊派蒙发送 `/map list` 查看资源列表", parse_mode="Markdown")
+            return
+        if resource_name in ("list", "列表"):
+            if filters.ChatType.GROUPS.filter(message):
+                reply_message = await message.reply_text("请私聊派蒙使用该命令")
+                self._add_delete_message_job(context, reply_message.chat_id, reply_message.message_id, 300)
+                self._add_delete_message_job(context, message.chat_id, message.message_id, 300)
+                return
+            Log.info(f"用户: {user.full_name} [{user.id}] 使用 map 命令查询了 资源列表")
+            text = self.map_helper.get_resource_list_mes()
+            await message.reply_text(text)
+            return
+        Log.info(f"用户: {user.full_name} [{user.id}] 使用 map 命令查询了 {resource_name}")
+        text = await self.map_helper.get_resource_map_mes(resource_name)
+        if "不知道" in text or "没有找到" in text:
+            await message.reply_text(text, parse_mode="Markdown")
+            return
+        img = Image.open(f"cache{sep}map.jpg")
+        if img.size[0] > 2048 or img.size[1] > 2048:
+            await message.reply_document(open(f"cache{sep}map.jpg", mode='rb+'), caption=text)
+        else:
+            await message.reply_photo(open(f"cache{sep}map.jpg", mode='rb+'), caption=text)

requirements.txt

@@ -16,4 +16,5 @@ lxml>=4.9.0
 fakeredis>=1.8.1
 aiohttp<=3.8.1
 python-telegram-bot==20.0a2
-pytz>=2021.3
+pytz>=2021.3
+Pillow

service/map.py

@@ -0,0 +1,313 @@
+import os
+import time
+from io import BytesIO
+from typing import Optional, List
+
+import httpx
+import ujson
+from PIL import Image, ImageMath
+
+from model.helpers import REQUEST_HEADERS
+
+Image.MAX_IMAGE_PIXELS = None
+
+ZOOM = 0.5
+RESOURCE_ICON_OFFSET = (-int(150 * 0.5 * ZOOM), -int(150 * ZOOM))
+
+
+class MapHelper:
+    LABEL_URL = 'https://api-static.mihoyo.com/common/blackboard/ys_obc/v1/map/label/tree?app_sn=ys_obc'
+    POINT_LIST_URL = 'https://api-static.mihoyo.com/common/blackboard/ys_obc/v1/map/point/list?map_id=2&app_sn=ys_obc'
+    MAP_URL = 'https://api-static.mihoyo.com/common/map_user/ys_obc/v1/map/info?map_id=2&app_sn=ys_obc&lang=zh-cn'
+
+    def __init__(self, cache_dir_name: str = "cache"):
+        self._current_dir = os.getcwd()
+        self._output_dir = os.path.join(self._current_dir, cache_dir_name)
+        self._resources_icon_dir = os.path.join(self._current_dir, "resources", "icon")
+        self._cache_dir = os.path.join(self._current_dir, "cache")
+        self._map_dir = os.path.join(self._cache_dir, "map_icon.jpg")
+        self.client = httpx.AsyncClient(headers=REQUEST_HEADERS, timeout=10.0)
+        self.all_resource_type: dict = {}
+        """这个字典保存所有资源类型
+
+        "1": {
+            "id": 1,
+            "name": "传送点",
+            "icon": "",
+            "parent_id": 0,
+            "depth": 1,
+            "node_type": 1,
+            "jump_type": 0,
+            "jump_target_id": 0,
+            "display_priority": 0,
+            "children": []
+        }
+        """
+        self.can_query_type_list: dict = {}
+        """这个字典保存所有可以查询的资源类型名称和ID，这个字典只有名称和ID
+
+        上边字典里"depth": 2的类型才可以查询，"depth": 1的是1级目录，不能查询
+        
+        "七天神像":"2"
+        
+        "风神瞳":"5"
+        """
+        self.all_resource_point_list: list = []
+        """这个列表保存所有资源点的数据
+
+        {
+            "id": 2740,
+            "label_id": 68,
+            "x_pos": -1789,
+            "y_pos": 2628,
+            "author_name": "作者名称",
+            "ctime": "更新时间",
+            "display_state": 1
+        }
+        """
+        self.date: str = ""
+        """记录上次更新"all_resource_point_list"的日期
+        """
+
+        self.center: Optional[List[float]] = None
+        """center
+        """
+
+        self.map_icon: Optional[Image] = None
+        """map_icon
+        """
+
+    async def download_icon(self, url):
+        """下载图片 返回Image对象
+        :param url:
+        :return:
+        """
+        resp = await self.client.get(url=url)
+        if resp.status_code != 200:
+            raise ValueError(f"获取图片数据失败，错误代码 {resp.status_code}")
+        icon = resp.content
+        return Image.open(BytesIO(icon))
+
+    async def download_json(self, url):
+        """
+        获取资源数据，返回 JSON
+        :param url:
+        :return: dict
+        """
+        resp = await self.client.get(url=url)
+        if resp.status_code != 200:
+            raise RuntimeError(f"获取资源点数据失败，错误代码 {resp.status_code}")
+        return resp.json()
+
+    async def init_point_list_and_map(self):
+        await self.up_label_and_point_list()
+        await self.up_map()
+
+    async def up_map(self):
+        """更新地图文件 并按照资源点的范围自动裁切掉不需要的地方
+        裁切地图需要最新的资源点位置，所以要先调用 up_label_and_point_list 再更新地图
+        :return: None
+        """
+        map_info = await self.download_json(self.MAP_URL)
+        map_info = map_info["data"]["info"]["detail"]
+        map_info = ujson.loads(map_info)
+
+        map_url_list = map_info['slices'][0]
+        origin = map_info["origin"]
+
+        x_start = map_info['total_size'][1]
+        y_start = map_info['total_size'][1]
+        x_end = 0
+        y_end = 0
+        for resource_point in self.all_resource_point_list:
+            x_pos = resource_point["x_pos"] + origin[0]
+            y_pos = resource_point["y_pos"] + origin[1]
+            x_start = min(x_start, x_pos)
+            y_start = min(y_start, y_pos)
+            x_end = max(x_end, x_pos)
+            y_end = max(y_end, y_pos)
+
+        x_start -= 200
+        y_start -= 200
+        x_end += 200
+        y_end += 200
+
+        self.center = [origin[0] - x_start, origin[1] - y_start]
+        x = int(x_end - x_start)
+        y = int(y_end - y_start)
+        self.map_icon = Image.new("RGB", (x, y))
+        x_offset = 0
+        for i in map_url_list:
+            map_url = i["url"]
+            map_icon = await self.download_icon(map_url)
+            self.map_icon.paste(map_icon, (int(-x_start) + x_offset, int(-y_start)))
+            x_offset += map_icon.size[0]
+
+    async def up_label_and_point_list(self):
+        """更新label列表和资源点列表
+        :return:
+        """
+        label_data = await self.download_json(self.LABEL_URL)
+        for label in label_data["data"]["tree"]:
+            self.all_resource_type[str(label["id"])] = label
+            for sublist in label["children"]:
+                self.all_resource_type[str(sublist["id"])] = sublist
+                self.can_query_type_list[sublist["name"]] = str(sublist["id"])
+                await self.up_icon_image(sublist)
+            label["children"] = []
+        test = await self.download_json(self.POINT_LIST_URL)
+        self.all_resource_point_list = test["data"]["point_list"]
+        self.date = time.strftime("%d")
+
+    async def up_icon_image(self, sublist: dict):
+        """检查是否有图标，没有图标下载保存到本地
+        :param sublist:
+        :return:
+        """
+        icon_id = sublist["id"]
+        icon_path = os.path.join(self._cache_dir, f"{icon_id}.png")
+
+        if not os.path.exists(icon_path):
+            icon_url = sublist["icon"]
+            icon = await self.download_icon(icon_url)
+            icon = icon.resize((150, 150))
+
+            box_alpha = Image.open(
+                os.path.join(os.path.dirname(__file__), os.path.pardir,
+                             "resources", "icon", "box_alpha.png")).getchannel("A")
+            box = Image.open(os.path.join(os.path.dirname(__file__), os.path.pardir, "resources", "icon", "box.png"))
+
+            try:
+                icon_alpha = icon.getchannel("A")
+                icon_alpha = ImageMath.eval("convert(a*b/256, 'L')", a=icon_alpha, b=box_alpha)
+            except ValueError:
+                # 米游社的图有时候会没有alpha导致报错，这时候直接使用box_alpha当做alpha就行
+                icon_alpha = box_alpha
+
+            icon2 = Image.new("RGBA", (150, 150), "#00000000")
+            icon2.paste(icon, (0, -10))
+
+            bg = Image.new("RGBA", (150, 150), "#00000000")
+            bg.paste(icon2, mask=icon_alpha)
+            bg.paste(box, mask=box)
+
+            with open(icon_path, "wb") as icon_file:
+                bg.save(icon_file)
+
+    async def get_resource_map_mes(self, name):
+        if self.date != time.strftime("%d"):
+            await self.init_point_list_and_map()
+        if name not in self.can_query_type_list:
+            return f"派蒙还不知道 {name} 在哪里呢，可以发送 `/map list` 查看资源列表"
+        resource_id = self.can_query_type_list[name]
+        map_res = ResourceMap(self.all_resource_point_list, self.map_icon, self.center, resource_id)
+        count = map_res.get_resource_count()
+        if not count:
+            return f"派蒙没有找到 {name} 的位置，可能米游社wiki还没更新"
+        map_res.gen_jpg()
+        return f"派蒙一共找到 {name} 的 {count} 个位置点\n* 数据来源于米游社wiki"
+
+    def get_resource_list_mes(self):
+        temp = {list_id: [] for list_id in self.all_resource_type if self.all_resource_type[list_id]["depth"] == 1}
+
+        for list_id in self.all_resource_type:
+            # 再找2级目录
+            if self.all_resource_type[list_id]["depth"] == 2:
+                temp[str(self.all_resource_type[list_id]["parent_id"])].append(list_id)
+        mes = "当前资源列表如下：\n"
+
+        for resource_type_id, value in temp.items():
+            if resource_type_id in ["1", "12", "50", "51", "95", "131"]:
+                # 在游戏里能查到的数据这里就不列举了，不然消息太长了
+                continue
+            mes += f"{self.all_resource_type[resource_type_id]['name']}："
+            for resource_id in value:
+                mes += f"{self.all_resource_type[resource_id]['name']}，"
+            mes += "\n"
+        return mes
+
+
+class ResourceMap:
+
+    def __init__(self, all_resource_point_list: List[dict], map_icon: Image, center: List[float], resource_id: int):
+        self.all_resource_point_list = all_resource_point_list
+        self.resource_id = resource_id
+        self.center = center
+        self.map_image = map_icon.copy()
+        self.map_size = self.map_image.size
+        # 地图要要裁切的左上角和右下角坐标
+        # 这里初始化为地图的大小
+        self.x_start = self.map_size[0]
+        self.y_start = self.map_size[1]
+        self.x_end = 0
+        self.y_end = 0
+        resource_icon = Image.open(self.get_icon_path())
+        self.resource_icon = resource_icon.resize((int(150 * ZOOM), int(150 * ZOOM)))
+        self.resource_xy_list = self.get_resource_point_list()
+
+    def get_icon_path(self):
+        # 检查有没有图标，有返回正确图标，没有返回默认图标
+        icon_path = os.path.join(os.path.dirname(__file__), os.path.pardir,
+                                 "resources", "icon", f"{self.resource_id}.png")
+        if os.path.exists(icon_path):
+            return icon_path
+        return os.path.join(os.path.dirname(__file__), os.path.pardir, "resources", "icon", "0.png")
+
+    def get_resource_point_list(self):
+        temp_list = []
+        for resource_point in self.all_resource_point_list:
+            if str(resource_point["label_id"]) == self.resource_id:
+                # 获取xy坐标，然后加上中心点的坐标完成坐标转换
+                x = resource_point["x_pos"] + self.center[0]
+                y = resource_point["y_pos"] + self.center[1]
+                temp_list.append((int(x), int(y)))
+        return temp_list
+
+    def paste(self):
+        for x, y in self.resource_xy_list:
+            # 把资源图片贴到地图上
+            # 这时地图已经裁切过了，要以裁切后的地图左上角为中心再转换一次坐标
+            x -= self.x_start
+            y -= self.y_start
+            self.map_image.paste(self.resource_icon, (x + RESOURCE_ICON_OFFSET[0], y + RESOURCE_ICON_OFFSET[1]),
+                                 self.resource_icon)
+
+    def crop(self):
+        # 把大地图裁切到只保留资源图标位置
+        for x, y in self.resource_xy_list:
+            # 找出4个方向最远的坐标，用于后边裁切
+            self.x_start = min(x, self.x_start)
+            self.y_start = min(y, self.y_start)
+            self.x_end = max(x, self.x_end)
+            self.y_end = max(y, self.y_end)
+
+        # 先把4个方向扩展150像素防止把资源图标裁掉
+        self.x_start -= 150
+        self.y_start -= 150
+        self.x_end += 150
+        self.y_end += 150
+
+        # 如果图片裁切得太小会看不出资源的位置在哪，检查图片裁切的长和宽看够不够1000，不到1000的按1000裁切
+        if (self.x_end - self.x_start) < 1000:
+            center = int((self.x_end + self.x_start) / 2)
+            self.x_start = center - 500
+            self.x_end = center + 500
+        if (self.y_end - self.y_start) < 1000:
+            center = int((self.y_end + self.y_start) / 2)
+            self.y_start = center - 500
+            self.y_end = center + 500
+
+        self.map_image = self.map_image.crop((self.x_start, self.y_start,
+                                              self.x_end, self.y_end))
+
+    def gen_jpg(self):
+        if not self.resource_xy_list:
+            return "没有这个资源的信息"
+        if not os.path.exists("cache"):
+            os.mkdir("cache")  # 查找 cache 目录 (缓存目录) 是否存在，如果不存在则创建
+        self.crop()
+        self.paste()
+        self.map_image.save(f'cache{os.sep}map.jpg', format='JPEG')
+
+    def get_resource_count(self):
+        return len(self.resource_xy_list)