Add HeXi bot codebase: custom plugins, web frontends, tests
- hexi core: message handling, rate limiting, cooldown, plugin manager - Custom plugins: BF stats, daily check-in, quotes, persona cards, etc. - Community plugins vendored under hexi/plugins with local fixes - Web admin frontends (learning-chat, persona-admin), unified hexi/web - Tests for rate_limit/cooldown/memes/persona; poetry.lock Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Claude
@@ -0,0 +1,76 @@
|
||||
# picfinder_take
|
||||
个人~~缝合~~修改的Hoshino bot搜图插件。
|
||||
|
||||
这个插件的主要思路其实是在hoshino上还原隔壁 [@Tsuk1ko](https://github.com/Tsuk1ko) 大佬家的[竹竹](https://github.com/Tsuk1ko/cq-picsearcher-bot)的搜图交互体验（所以插件名是たけ）
|
||||
|
||||
代码主体部分参考了 [@Watanabe-Asa](https://github.com/Watanabe-Asa)大佬的 [搜图](https://github.com/pcrbot/Salmon-plugin-transplant#%E6%90%9C%E5%9B%BE)与 [@Cappuccilo](https://github.com/Cappuccilo)大佬的 [以图搜图](https://github.com/pcrbot/cappuccilo_plugins#%E4%BB%A5%E5%9B%BE%E6%90%9C%E5%9B%BE)，感谢各位大佬的代码）
|
||||
|
||||
---
|
||||
|
||||
- 6.1更新：增加私聊搜图、截屏识别和代理功能
|
||||
|
||||
- 6.24更新：增加回复搜图功能，搜图请求更换为异步（感谢 [@蓝红心](https://github.com/LHXnois)大佬）
|
||||
|
||||
- 7.9更新：增加自定义HOST功能，在qiang内又不想用全局代理的用户(?)可以单独为SauceNao和ascii2d配置反代或ip直连）（继续感谢 [@蓝红心](https://github.com/LHXnois)大佬）
|
||||
|
||||
- 7.26更新：回复搜图增加at支持
|
||||
|
||||
- 10.20更新：增加批量搜图计数与吞图提示
|
||||
|
||||
- 12.12更新：增加频道搜图支持，适配go-cqhttp-1.0.0beat8-fix2（还是感谢 [@蓝红心](https://github.com/LHXnois)大佬）
|
||||
|
||||
- 22.01.10更新：适配新版图片CQcode，增加IGNORE_STAMP配置项，可在批量搜索中自动忽略subType不为0的图片（比如表情包）；用cloudscraper绕过ascii2d近期新增的cf认证
|
||||
|
||||
## 特点
|
||||
|
||||
- 搜索SauceNao，在相似率过低时自动补充搜索ascii2d，相似率阈值可在config中调整。搜索结果显示数量可在config中调整。
|
||||
|
||||
- 解析SauceNao和ascii2d搜索结果的作品详细信息。SauceNao全部42个index格式解析都已完成；ascii2d常见格式应该也能解析，一些奇怪格式的外部登录不敢保证）
|
||||
|
||||
- 获取SauceNao和ascii2d的结果缩略图，缩略图可在config中关闭。
|
||||
|
||||
- 搜图结果可由普通回复切换为合并转发回复，减少刷屏情况。发送模式可在config中调整。
|
||||
|
||||
- 增加批量搜图模式，解决移动端文字命令+图片发送麻烦的问题。
|
||||
|
||||
- 增加搜图每人每日限额，可在config中调整限额。要懂得节制噢.jpg
|
||||
|
||||
- 增加简单的手机截屏识别功能，判断为整屏手机截屏时会拒绝搜索 ~~（你会截你马个图.jpg）~~
|
||||
|
||||
- 增加私聊搜图功能，有效缓解腾讯吞图（但反之临时会话下搜索结果含图片与链接时概率被吞无法发送，若要稳定使用需加bot好友）
|
||||
|
||||
- 增加代理设置，方便qiang内使用
|
||||
|
||||
- 增加回复搜图功能，可直接对群友发送的图片进行回复搜索，省去转发过程）
|
||||
|
||||
- New！增加IGNORE_STAMP配置项，可在批量搜索中自动忽略沙雕群友发的表情包，减少资源浪费与刷屏）
|
||||
|
||||
|
||||
## 用法
|
||||
|
||||
- 申请并在config中配置SauceNao的API key
|
||||
|
||||
- 发送 ``@bot+图片`` 或 ``[bot昵称]搜图+图片`` 进行单张搜索。
|
||||
|
||||
- 发送 ``[bot昵称]搜图`` 进入连续搜图模式，连续搜图模式下同一用户所发送所有图片都将直接搜索；
|
||||
|
||||
发送 ``谢谢[bot昵称]`` 退出连续搜图模式，或停止发图等待超时后自动退出连续搜图模式。
|
||||
|
||||
- 对他人发送的图片回复 ``@bot 搜图`` 或``[bot昵称]搜图`` 进行回复搜索。
|
||||
|
||||
- 私聊下直接发送图片即可进行搜索。
|
||||
|
||||
|
||||
## 频道使用
|
||||
|
||||
对特定子频道中收到的所有图片进行搜索（本质应该算是个常驻的连续搜图？）
|
||||
在qq群这样可能影响讨论，但是在频道可以专门开一个搜图子频道只用来搜图！
|
||||
|
||||
食用方法：
|
||||
|
||||
- 创建一个专门的子频道
|
||||
- 搞到频道id和子频道id（看log）
|
||||
- 按{频道id: [子频道id]}格式填到config.py里的enableguild里
|
||||
- 发送图片开始搜图
|
||||
|
||||
开启慢速模式然后给bot管理体验更佳
|
||||
@@ -0,0 +1,466 @@
|
||||
import re
|
||||
from asyncio import sleep
|
||||
from datetime import datetime, timedelta
|
||||
|
||||
from nonebot import get_driver, on_message, on_startswith, require
|
||||
from nonebot.adapters.onebot.v11 import Bot, GroupMessageEvent, PrivateMessageEvent
|
||||
from nonebot.exception import ActionFailed
|
||||
from nonebot.log import logger
|
||||
from nonebot.plugin import PluginMetadata
|
||||
from nonebot.rule import is_type, to_me
|
||||
|
||||
require("nonebot_plugin_alconna")
|
||||
from nonebot_plugin_alconna import UniMessage
|
||||
|
||||
from .config import (
|
||||
CHAIN_REPLY,
|
||||
CHECK,
|
||||
DAILY_LIMIT,
|
||||
IGNORE_STAMP,
|
||||
SAUCENAO_KEY,
|
||||
SEARCH_TIMEOUT,
|
||||
threshold,
|
||||
)
|
||||
from .image import check_screenshot, get_image_data_ascii, get_image_data_sauce
|
||||
|
||||
# 统一配置标准：把「插件内部常量」也暴露给 Web 管理台（来源无关）。
|
||||
# 说明：Web 读取/修改会写入 plugin_config.json 并热更新 config 模块属性；
|
||||
# 要在运行期立即生效，插件各使用点应改用 get_effective_value 读取（见 docs/plugin-config-standard.md §5b）。
|
||||
from hexi.config_standard import register_config_items # noqa: E402
|
||||
from . import config as _cfgmod # noqa: E402
|
||||
|
||||
register_config_items(
|
||||
__name__,
|
||||
[
|
||||
{"key": "SAUCENAO_KEY", "label": "SauceNAO API Key", "type": "password", "secret": True},
|
||||
{"key": "DAILY_LIMIT", "label": "每日搜图限额", "type": "int", "default": 50},
|
||||
{"key": "SEARCH_TIMEOUT", "label": "批量搜索超时(秒)", "type": "int", "default": 60},
|
||||
{"key": "CHECK", "label": "开启截屏判定", "type": "bool", "default": True},
|
||||
{"key": "THUMB_ON", "label": "启用缩略图", "type": "bool", "default": True},
|
||||
{"key": "CHAIN_REPLY", "label": "合并转发回复", "type": "bool", "default": True},
|
||||
{"key": "IGNORE_STAMP", "label": "批量忽略表情包", "type": "bool", "default": True},
|
||||
{"key": "threshold", "label": "相似度阈值", "type": "float", "default": 70},
|
||||
],
|
||||
store=_cfgmod,
|
||||
)
|
||||
|
||||
__plugin_meta__ = PluginMetadata(
|
||||
name="搜图",
|
||||
description="SauceNAO + ascii2d 二次元搜图",
|
||||
usage=(
|
||||
"(xx为bot名称)\n"
|
||||
"[@bot+图片] 单张/多张搜图\n"
|
||||
"[xx搜图] 进入批量搜图模式\n"
|
||||
"[谢谢xx] 退出批量搜图模式"
|
||||
),
|
||||
type="application",
|
||||
)
|
||||
|
||||
driver_config = get_driver().config
|
||||
SUPERUSERS = driver_config.superusers
|
||||
NICKNAMES = list(driver_config.nickname) if driver_config.nickname else ["竹竹"]
|
||||
NICKNAME = NICKNAMES[0]
|
||||
|
||||
class DailyNumberLimiter:
|
||||
"""每日次数限制器(按年积日计数)"""
|
||||
|
||||
def __init__(self, max_num):
|
||||
self.max = max_num
|
||||
self.today = -1
|
||||
self.count = {}
|
||||
|
||||
def check(self, key) -> bool:
|
||||
day = datetime.now().timetuple().tm_yday
|
||||
if day != self.today:
|
||||
self.today = day
|
||||
self.count.clear()
|
||||
return self.count.get(key, 0) < self.max
|
||||
|
||||
def get_num(self, key):
|
||||
return self.count.get(key, 0)
|
||||
|
||||
def increase(self, key, num=1):
|
||||
self.count[key] = self.count.get(key, 0) + num
|
||||
|
||||
|
||||
lmtd = DailyNumberLimiter(DAILY_LIMIT)
|
||||
|
||||
|
||||
class PicListener:
|
||||
def __init__(self):
|
||||
self.on = {}
|
||||
self.count = {}
|
||||
self.limit = {}
|
||||
self.timeout = {}
|
||||
|
||||
def get_on_off_status(self, gid):
|
||||
return self.on[gid] if self.on.get(gid) is not None else False
|
||||
|
||||
def turn_on(self, gid, uid):
|
||||
self.on[gid] = uid
|
||||
self.timeout[gid] = datetime.now() + timedelta(seconds=SEARCH_TIMEOUT)
|
||||
self.count[gid] = 0
|
||||
self.limit[gid] = DAILY_LIMIT - lmtd.get_num(uid)
|
||||
|
||||
def turn_off(self, gid):
|
||||
self.on.pop(gid)
|
||||
self.count.pop(gid)
|
||||
self.timeout.pop(gid)
|
||||
self.limit.pop(gid)
|
||||
|
||||
def count_plus(self, gid):
|
||||
self.count[gid] += 1
|
||||
|
||||
|
||||
pls = PicListener()
|
||||
|
||||
start_finder = on_startswith(("识图", "搜图", "查图", "找图"), rule=to_me())
|
||||
picmessage = on_message(rule=is_type(GroupMessageEvent), block=False)
|
||||
replymessage = on_message(rule=is_type(GroupMessageEvent), block=False)
|
||||
thanks = on_startswith("谢谢")
|
||||
pic_private = on_message(rule=is_type(PrivateMessageEvent), block=False)
|
||||
|
||||
|
||||
def parse_report(text: str) -> UniMessage:
|
||||
"""将含 [CQ:image,file=...] 的搜索报告字符串转为 UniMessage"""
|
||||
msg = UniMessage()
|
||||
parts = re.split(r"(\[CQ:image,[^\]]*\])", text)
|
||||
for part in parts:
|
||||
if part.startswith("[CQ:image,"):
|
||||
m = re.search(r"file=([^,\]]+)", part)
|
||||
msg += UniMessage.image(url=m.group(1) if m else "")
|
||||
elif part:
|
||||
msg += UniMessage.text(part)
|
||||
return msg
|
||||
|
||||
|
||||
def _extract_image(ev) -> tuple[str, str, str | None] | None:
|
||||
"""从消息中提取第一张图片的 (file, url, subType),无图片返回 None"""
|
||||
for m in ev.message:
|
||||
if m.type == "image":
|
||||
return m.data["file"], m.data["url"], m.data.get("subType")
|
||||
return None
|
||||
|
||||
|
||||
@start_finder.handle()
|
||||
async def start_finder_handle(bot: Bot, ev: GroupMessageEvent):
|
||||
uid = ev.user_id
|
||||
gid = ev.group_id
|
||||
mid = ev.message_id
|
||||
if str(uid) not in SUPERUSERS:
|
||||
await UniMessage.text("暂不支持在群聊中使用！").send()
|
||||
return
|
||||
ret = _extract_image(ev)
|
||||
if not ret:
|
||||
if pls.get_on_off_status(gid):
|
||||
if uid == pls.on[gid]:
|
||||
pls.timeout[gid] = datetime.now() + timedelta(seconds=30)
|
||||
await UniMessage.text(
|
||||
"您已经在搜图模式下啦！\n如想退出搜图模式请发送“谢谢竹竹”~"
|
||||
).send()
|
||||
await start_finder.finish()
|
||||
else:
|
||||
await UniMessage.at(user_id=str(pls.on[gid])).text(
|
||||
"正在搜图，请耐心等待~"
|
||||
).send()
|
||||
await start_finder.finish()
|
||||
pls.turn_on(gid, uid)
|
||||
await UniMessage.text(
|
||||
f"了解～请发送图片吧！支持批量噢！\n如想退出搜索模式请发送“谢谢{NICKNAME}”"
|
||||
).send()
|
||||
await sleep(30)
|
||||
ct = 0
|
||||
while pls.get_on_off_status(gid):
|
||||
if datetime.now() < pls.timeout[gid]:
|
||||
if ct != pls.count[gid]:
|
||||
ct = pls.count[gid]
|
||||
pls.timeout[gid] = datetime.now() + timedelta(seconds=60)
|
||||
else:
|
||||
temp = pls.on[gid]
|
||||
if not pls.count[gid]:
|
||||
await UniMessage.at(user_id=str(temp)).text(
|
||||
" 由于超时，已为您自动退出搜图模式~\n您本次搜索期间未发送任何图片，请检查是否被吞图~"
|
||||
).send()
|
||||
else:
|
||||
await UniMessage.at(user_id=str(temp)).text(
|
||||
f" 由于超时，已为您自动退出搜图模式，以后要记得说“谢谢{NICKNAME}”来退出搜图模式噢~\n您本次搜索共搜索了{pls.count[gid]}张图片～"
|
||||
).send()
|
||||
pls.turn_off(ev.group_id)
|
||||
break
|
||||
await sleep(30)
|
||||
return
|
||||
file, url, _ = ret
|
||||
if str(uid) not in SUPERUSERS:
|
||||
if not lmtd.check(uid):
|
||||
await UniMessage.text(
|
||||
f"您今天已经搜过{DAILY_LIMIT}次图了，休息一下明天再来吧~"
|
||||
).send(at_sender=True)
|
||||
return
|
||||
|
||||
if CHECK:
|
||||
result = await check_screenshot(bot, file, url)
|
||||
if result:
|
||||
if result == 1:
|
||||
await UniMessage.reply(id=str(mid)).text(
|
||||
"该图似乎是手机截屏，请进行适当裁剪后再尝试搜图~\n*请注意搜索漫画时务必截取一个完整单页进行搜图~"
|
||||
).send()
|
||||
if result == 2:
|
||||
await UniMessage.reply(id=str(mid)).text(
|
||||
"该图似乎是长图拼接，请进行适当裁剪后再尝试搜图~\n*请注意搜索漫画时务必截取一个完整单页进行搜图~"
|
||||
).send()
|
||||
return
|
||||
await UniMessage.text("正在搜索，请稍候～").send()
|
||||
await picfinder(bot, ev, url)
|
||||
|
||||
|
||||
@picmessage.handle()
|
||||
async def picmessage_handle(bot: Bot, ev: GroupMessageEvent):
|
||||
mid = ev.message_id
|
||||
atcheck = False
|
||||
batchcheck = False
|
||||
for m in ev.message:
|
||||
if m.type == "at" and int(m.data["qq"]) == ev.self_id:
|
||||
atcheck = True
|
||||
if pls.get_on_off_status(ev.group_id):
|
||||
if int(pls.on[ev.group_id]) == int(ev.user_id):
|
||||
batchcheck = True
|
||||
if not (batchcheck or atcheck):
|
||||
return
|
||||
uid = ev.user_id
|
||||
ret = _extract_image(ev)
|
||||
if not ret:
|
||||
return
|
||||
file, url, sbtype = ret
|
||||
if str(uid) not in SUPERUSERS:
|
||||
if not lmtd.check(uid):
|
||||
await UniMessage.text(
|
||||
f"您今天已经搜过{DAILY_LIMIT}次图了，休息一下明天再来吧～"
|
||||
).send(at_sender=True)
|
||||
if pls.get_on_off_status(ev.group_id):
|
||||
pls.turn_off(ev.group_id)
|
||||
return
|
||||
if pls.get_on_off_status(ev.group_id):
|
||||
pls.count_plus(ev.group_id)
|
||||
if pls.count[ev.group_id] > pls.limit[ev.group_id]:
|
||||
await UniMessage.text(
|
||||
f"您今天已经搜过{DAILY_LIMIT}次图了，休息一下明天再来吧～"
|
||||
).send(at_sender=True)
|
||||
pls.turn_off(ev.group_id)
|
||||
return
|
||||
if sbtype and IGNORE_STAMP:
|
||||
if sbtype != "0":
|
||||
await UniMessage.reply(id=str(mid)).text(
|
||||
"该图为表情，已忽略~如确需搜索请尝试单发搜索或回复搜索~"
|
||||
).send()
|
||||
return
|
||||
|
||||
if CHECK:
|
||||
result = await check_screenshot(bot, file, url)
|
||||
if result:
|
||||
if result == 1:
|
||||
await UniMessage.reply(id=str(mid)).text(
|
||||
"该图似乎是手机截屏，请手动进行适当裁剪后再尝试搜图~\n*请注意搜索漫画时务必截取一个完整单页进行搜图~"
|
||||
).send()
|
||||
if result == 2:
|
||||
await UniMessage.reply(id=str(mid)).text(
|
||||
"该图似乎是长图拼接，请手动进行适当裁剪后再尝试搜图~\n*请注意搜索漫画时务必截取一个完整单页进行搜图~"
|
||||
).send()
|
||||
return
|
||||
if "c2cpicdw.qpic.cn/offpic_new/" in url:
|
||||
md5 = file[:-6].upper()
|
||||
url = f"http://gchat.qpic.cn/gchatpic_new/0/0-0-{md5}/0?term=2"
|
||||
await UniMessage.text("正在搜索，请稍候～").send()
|
||||
await picfinder(bot, ev, url)
|
||||
|
||||
|
||||
@replymessage.handle()
|
||||
async def replymessage_handle(bot: Bot, ev: GroupMessageEvent):
|
||||
mid = ev.message_id
|
||||
uid = ev.user_id
|
||||
seg = ev.message[0]
|
||||
if seg.type != "reply":
|
||||
return
|
||||
tmid = seg.data["id"]
|
||||
cmd = ev.message.extract_plain_text()
|
||||
flag1 = 0
|
||||
flag2 = 0
|
||||
for m in ev.message[2:]:
|
||||
if m.type == "at" and int(m.data["qq"]) == ev.self_id:
|
||||
flag1 = 1
|
||||
for name in NICKNAMES:
|
||||
if name in cmd:
|
||||
flag1 = 1
|
||||
break
|
||||
for pfcmd in ["识图", "搜图", "查图", "找图"]:
|
||||
if pfcmd in cmd:
|
||||
flag2 = 1
|
||||
if not (flag1 and flag2):
|
||||
return
|
||||
if str(uid) not in SUPERUSERS:
|
||||
if not lmtd.check(uid):
|
||||
await UniMessage.text(
|
||||
f"您今天已经搜过{DAILY_LIMIT}次图了，休息一下明天再来吧～"
|
||||
).send(at_sender=True)
|
||||
try:
|
||||
tmsg = await bot.get_msg(message_id=int(tmid))
|
||||
except ActionFailed:
|
||||
await UniMessage.text("该消息已过期，请重新转发~").send()
|
||||
await replymessage.finish()
|
||||
ret = re.search(r"\[CQ:image,file=(.*)?,url=(.*)\]", str(tmsg["message"]))
|
||||
if not ret:
|
||||
await UniMessage.text("未找到图片~").send()
|
||||
return
|
||||
file = ret.group(1)
|
||||
url = ret.group(2)
|
||||
|
||||
if ",subType=" in url:
|
||||
url = url.split(",")[0]
|
||||
elif ",subType=" in file:
|
||||
file = file.split(",")[0]
|
||||
|
||||
if CHECK:
|
||||
result = await check_screenshot(bot, file, url)
|
||||
if result:
|
||||
if result == 1:
|
||||
await UniMessage.reply(id=str(mid)).text(
|
||||
"该图似乎是手机截屏，请手动进行适当裁剪后再尝试搜图~\n*请注意搜索漫画时务必截取一个完整单页进行搜图~"
|
||||
).send()
|
||||
if result == 2:
|
||||
await UniMessage.reply(id=str(mid)).text(
|
||||
"该图似乎是长图拼接，请手动进行适当裁剪后再尝试搜图~\n*请注意搜索漫画时务必截取一个完整单页进行搜图~"
|
||||
).send()
|
||||
return
|
||||
if "c2cpicdw.qpic.cn/offpic_new/" in url:
|
||||
md5 = file[:-6].upper()
|
||||
url = f"http://gchat.qpic.cn/gchatpic_new/0/0-0-{md5}/0?term=2"
|
||||
await UniMessage.text("正在搜索，请稍候～").send()
|
||||
await picfinder(bot, ev, url)
|
||||
|
||||
|
||||
@thanks.handle()
|
||||
async def thanks_handle(bot: Bot, ev: GroupMessageEvent):
|
||||
gid = ev.group_id
|
||||
name = ev.message.extract_plain_text().removeprefix("谢谢").strip()
|
||||
if name not in NICKNAMES:
|
||||
return
|
||||
if pls.get_on_off_status(gid):
|
||||
if pls.on[gid] != ev.user_id:
|
||||
await UniMessage.text("不能替别人结束搜图哦～").send()
|
||||
return
|
||||
if not pls.count[gid]:
|
||||
await UniMessage.text("不用谢～\n您本次搜索期间未发送任何图片，请检查是否被吞图～").send()
|
||||
else:
|
||||
await UniMessage.text(f"不用谢～\n您本次搜索共搜索了{pls.count[gid]}张图片～").send()
|
||||
pls.turn_off(gid)
|
||||
return
|
||||
await UniMessage.text("にゃ～").send()
|
||||
|
||||
|
||||
@pic_private.handle()
|
||||
async def pic_private_handle(bot: Bot, ev: PrivateMessageEvent):
|
||||
uid = ev.user_id
|
||||
ret = _extract_image(ev)
|
||||
if not ret:
|
||||
flag1 = flag2 = 0
|
||||
for name in NICKNAMES:
|
||||
if name in str(ev.message):
|
||||
flag1 = 1
|
||||
break
|
||||
for pfcmd in ["识图", "搜图", "查图", "找图"]:
|
||||
if pfcmd in str(ev.message):
|
||||
flag2 = 1
|
||||
if flag1 and flag2:
|
||||
await UniMessage.text("私聊搜图请直接发送图片~").send()
|
||||
return
|
||||
file, url, _ = ret
|
||||
if not lmtd.check(uid):
|
||||
await UniMessage.text(
|
||||
f"您今天已经搜过{DAILY_LIMIT}次图了，休息一下明天再来吧~"
|
||||
).send()
|
||||
return
|
||||
if "c2cpicdw.qpic.cn/offpic_new/" in url:
|
||||
md5 = file.upper()
|
||||
logger.info(f"图片URL{url}")
|
||||
url = f"https://gchat.qpic.cn/gchatpic_new/0/0-0-{md5}/0?term=2"
|
||||
|
||||
await UniMessage.text("正在搜索，请稍候～").send()
|
||||
result = await get_image_data_sauce(url, SAUCENAO_KEY)
|
||||
image_data_report = result[0]
|
||||
simimax = result[1]
|
||||
if "Index #" in image_data_report:
|
||||
su = next(iter(SUPERUSERS))
|
||||
await bot.send_private_msg(user_id=int(su), message="发生index解析错误")
|
||||
await bot.send_private_msg(user_id=int(su), message=url)
|
||||
await bot.send_private_msg(user_id=int(su), message=image_data_report)
|
||||
await parse_report(image_data_report).send()
|
||||
|
||||
if float(simimax) > float(threshold):
|
||||
lmtd.increase(uid)
|
||||
else:
|
||||
if simimax != 0:
|
||||
await UniMessage.text("相似度过低，换用ascii2d检索中…").send()
|
||||
else:
|
||||
logger.error("SauceNao not found imageInfo")
|
||||
await UniMessage.text("SauceNao检索失败,换用ascii2d检索中…").send()
|
||||
|
||||
image_data_report = await get_image_data_ascii(url)
|
||||
if image_data_report[0]:
|
||||
await parse_report(image_data_report[0]).send()
|
||||
lmtd.increase(uid)
|
||||
if image_data_report[1]:
|
||||
await parse_report(image_data_report[1]).send()
|
||||
if not (image_data_report[0] or image_data_report[1]):
|
||||
logger.error("ascii2d not found imageInfo")
|
||||
await UniMessage.text("ascii2d检索失败…").send()
|
||||
|
||||
|
||||
async def chain_reply(bot: Bot, ev: GroupMessageEvent, chain, msg):
|
||||
if not CHAIN_REPLY:
|
||||
await parse_report(msg).send()
|
||||
return chain
|
||||
data = {
|
||||
"type": "node",
|
||||
"data": {
|
||||
"name": str(NICKNAME) if str(NICKNAME) else "竹竹",
|
||||
"uin": str(ev.self_id),
|
||||
"content": str(msg),
|
||||
},
|
||||
}
|
||||
chain.append(data)
|
||||
return chain
|
||||
|
||||
|
||||
async def picfinder(bot: Bot, ev: GroupMessageEvent, image_data):
|
||||
uid = ev.user_id
|
||||
chain = []
|
||||
result = await get_image_data_sauce(image_data, SAUCENAO_KEY)
|
||||
image_data_report = result[0]
|
||||
simimax = result[1]
|
||||
if "Index #" in image_data_report:
|
||||
su = next(iter(SUPERUSERS))
|
||||
await bot.send_private_msg(user_id=int(su), message="发生index解析错误")
|
||||
await bot.send_private_msg(user_id=int(su), message=image_data)
|
||||
await bot.send_private_msg(user_id=int(su), message=image_data_report)
|
||||
chain = await chain_reply(bot, ev, chain, image_data_report)
|
||||
|
||||
if float(simimax) > float(threshold):
|
||||
lmtd.increase(uid)
|
||||
else:
|
||||
if simimax != 0:
|
||||
chain = await chain_reply(bot, ev, chain, "相似度过低，换用ascii2d检索中…")
|
||||
else:
|
||||
logger.error("SauceNao not found imageInfo")
|
||||
chain = await chain_reply(bot, ev, chain, "SauceNao检索失败,换用ascii2d检索中…")
|
||||
|
||||
image_data_report = await get_image_data_ascii(image_data)
|
||||
if image_data_report[0]:
|
||||
chain = await chain_reply(bot, ev, chain, image_data_report[0])
|
||||
lmtd.increase(uid)
|
||||
if image_data_report[1]:
|
||||
chain = await chain_reply(bot, ev, chain, image_data_report[1])
|
||||
if not (image_data_report[0] or image_data_report[1]):
|
||||
logger.error("ascii2d not found imageInfo")
|
||||
chain = await chain_reply(bot, ev, chain, "ascii2d检索失败…")
|
||||
|
||||
if CHAIN_REPLY:
|
||||
await bot.send_group_forward_msg(group_id=ev.group_id, messages=chain)
|
||||
@@ -0,0 +1,50 @@
|
||||
import os
|
||||
|
||||
threshold = 70 # SauceNAO相似度阈值，低于该相似度自动追加ascii2d搜索
|
||||
|
||||
SAUCENAO_KEY = "dfc2b9156f254754dee2ead531826020f6316f92" # SauceNAO 的 API key
|
||||
|
||||
SAUCENAO_RESULT_NUM = 3 # SauceNAO搜索结果显示数量
|
||||
|
||||
ASCII_RESULT_NUM = 3 # ascii2d搜索结果显示数量
|
||||
|
||||
SEARCH_TIMEOUT = 60 # 连续搜索模式超时时长
|
||||
|
||||
DAILY_LIMIT = 50 # 搜图每日限额
|
||||
|
||||
CHAIN_REPLY = True # 是否启用合并转发回复模式
|
||||
|
||||
THUMB_ON = True # 是否启用缩略图
|
||||
|
||||
CHECK = True # 是否开启手机截屏判定
|
||||
|
||||
IGNORE_STAMP = True # 是否在批量搜索中忽略表情包
|
||||
|
||||
HOST_CUSTOM = {
|
||||
# 自定义Host，不使用留空即可
|
||||
# 格式示例：'https://ascii2d.net' , 'http://localhost:12345'
|
||||
'SAUCENAO': '',
|
||||
'ASCII': ''
|
||||
}
|
||||
|
||||
# ascii2d 的 Cloudflare cookie 文件(Netscape 格式,浏览器导出,参考 nonebot_plugin_video_analysis 的 cookies.txt 用法)
|
||||
# 获取方式:浏览器打开 ascii2d.net 过验证后,用 EditThisCookie / Cookie-Editor 扩展导出为 Netscape 格式,
|
||||
# 保存为 nonebot_plugin_picfinder_take/data/cookies_ascii2d.txt
|
||||
# 留空则使用 playwright 无 cookie 尝试(可能无法通过 Cloudflare 验证)
|
||||
ASCII2D_COOKIES_FILE = os.path.join(
|
||||
os.path.dirname(os.path.abspath(__file__)), "data", "cookies_ascii2d.txt"
|
||||
)
|
||||
|
||||
proxies = {
|
||||
'http': '',
|
||||
'https': ''
|
||||
} # 网络代理
|
||||
|
||||
enableguild = {0000: [00000, 1111], 123456: [2345]} # 频道白名单 格式{频道id: [子频道id]}
|
||||
|
||||
helptext = '''
|
||||
(xx为bot名称)
|
||||
[@bot+图片] 单张/多张搜图
|
||||
[xx搜图] 进入批量搜图模式
|
||||
[谢谢xx] 退出批量搜图模式
|
||||
'''.strip()
|
||||
@@ -0,0 +1,865 @@
|
||||
import asyncio
|
||||
import base64
|
||||
import os
|
||||
import re
|
||||
from io import BytesIO
|
||||
from random import randint
|
||||
from traceback import format_exc
|
||||
|
||||
import cloudscraper
|
||||
import httpx
|
||||
from PIL import Image
|
||||
from lxml import etree
|
||||
from nonebot.log import logger
|
||||
from playwright.async_api import async_playwright
|
||||
|
||||
from .config import (
|
||||
ASCII2D_COOKIES_FILE,
|
||||
ASCII_RESULT_NUM,
|
||||
HOST_CUSTOM,
|
||||
SAUCENAO_RESULT_NUM,
|
||||
THUMB_ON,
|
||||
proxies,
|
||||
)
|
||||
|
||||
|
||||
def pic2b64(pic) -> str:
|
||||
"""PIL Image 转 base64:// 字符串"""
|
||||
buf = BytesIO()
|
||||
pic.save(buf, format="PNG")
|
||||
return "base64://" + base64.b64encode(buf.getvalue()).decode()
|
||||
|
||||
|
||||
def _img_cq(base64_str: str) -> str:
|
||||
"""生成图片 CQ 码字符串(用于嵌入搜索报告文本)"""
|
||||
return f"[CQ:image,file={base64_str}]"
|
||||
|
||||
|
||||
def _make_client(timeout: float) -> httpx.AsyncClient:
|
||||
"""构造 httpx 客户端(httpx 0.28 移除了 AsyncClient 的 proxies 参数,转为 mounts)"""
|
||||
http_proxy = proxies.get("http")
|
||||
https_proxy = proxies.get("https")
|
||||
if http_proxy or https_proxy:
|
||||
mounts = {}
|
||||
if http_proxy:
|
||||
mounts["http://"] = httpx.HTTPTransport(proxy=http_proxy)
|
||||
if https_proxy:
|
||||
mounts["https://"] = httpx.HTTPTransport(proxy=https_proxy)
|
||||
return httpx.AsyncClient(mounts=mounts, timeout=timeout)
|
||||
return httpx.AsyncClient(timeout=timeout)
|
||||
|
||||
|
||||
async def get_pic(address):
|
||||
async with _make_client(20) as client:
|
||||
resp = await client.get(address)
|
||||
return resp.content
|
||||
|
||||
|
||||
def randcolor():
|
||||
return (randint(0, 255), randint(0, 255), randint(0, 255))
|
||||
|
||||
|
||||
def ats_pic(img):
|
||||
if img.mode != "RGB":
|
||||
img = img.convert("RGB")
|
||||
width = img.size[0] - 1 # 长度
|
||||
height = img.size[1] - 1 # 宽度
|
||||
img.putpixel((0, 0), randcolor())
|
||||
img.putpixel((0, height), randcolor())
|
||||
img.putpixel((width, 0), randcolor())
|
||||
img.putpixel((width, height), randcolor())
|
||||
return img
|
||||
|
||||
|
||||
async def check_screenshot(bot, file, imgurl):
|
||||
async with _make_client(20) as client:
|
||||
pichead = await client.head(imgurl)
|
||||
if pichead.headers["Content-Type"] == "image/gif":
|
||||
print("gif pic, not likely a screen shot")
|
||||
return 0
|
||||
try:
|
||||
resp = await client.get(imgurl)
|
||||
image = Image.open(BytesIO(resp.content))
|
||||
except Exception:
|
||||
print("download failed")
|
||||
return 0
|
||||
cord = image.size[0] / image.size[1]
|
||||
height = image.size[1]
|
||||
print(cord)
|
||||
if cord > 0.565:
|
||||
print("too short, not likely a screen shot")
|
||||
return 0
|
||||
if cord < 0.2:
|
||||
print("too long, might be long screen shot")
|
||||
return 2
|
||||
print("size checked, next ocr")
|
||||
try:
|
||||
ocr_result = await bot.call_api(".ocr_image", image=file)
|
||||
except Exception:
|
||||
print("ocr failed")
|
||||
return False
|
||||
flag = 0
|
||||
for result in ocr_result["texts"]:
|
||||
key1 = re.search("[0-9]{1,2}:[0-9]{2}", result["text"]) # 时间
|
||||
key2 = re.search("移动|联通|电信", result["text"])
|
||||
key3 = re.search("4G|5G", result["text"])
|
||||
key4 = re.search("[0-9]{1,2}%", result["text"]) # 电量
|
||||
key5 = re.search("[0-9]{0,3}[\\\/][0-9]{0,3}", result["text"]) # 页数
|
||||
if key2 or key3 or key4:
|
||||
print(str(result))
|
||||
loc = result["coordinates"][2]["y"]
|
||||
if int(loc) < (int(height) / 19):
|
||||
flag = 1
|
||||
if key1 or key5:
|
||||
print(str(result))
|
||||
loc = result["coordinates"][2]["y"]
|
||||
if int(loc) < (int(height) / 19) or int(loc) > (int(height) * 18 / 20):
|
||||
flag = 1
|
||||
if flag:
|
||||
break
|
||||
if flag:
|
||||
return 1
|
||||
else:
|
||||
return 0
|
||||
|
||||
|
||||
def sauces_info(sauce):
|
||||
service_name = ""
|
||||
info = ""
|
||||
try:
|
||||
if sauce["header"]["index_id"] == 0:
|
||||
service_name = "H-Magazines"
|
||||
title = sauce["data"]["title"]
|
||||
part = sauce["data"]["part"]
|
||||
date = sauce["data"]["date"]
|
||||
info = f"{title}-{part}/{date}"
|
||||
|
||||
# index 1 "h-anime" disabled
|
||||
|
||||
elif sauce["header"]["index_id"] == 2:
|
||||
service_name = "H-Game CG"
|
||||
company = sauce["data"]["company"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"[{company}] {title}"
|
||||
|
||||
# index 3 "ddb-objects" disabled
|
||||
|
||||
# index 4 "ddb-samples" disabled
|
||||
|
||||
elif sauce["header"]["index_id"] == 5 or sauce["header"]["index_id"] == 6:
|
||||
service_name = "pixiv"
|
||||
author_name = sauce["data"]["member_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
# index 6 "pixiv historical" with 5
|
||||
|
||||
# index 7 "anime" disabled
|
||||
|
||||
elif sauce["header"]["index_id"] == 8:
|
||||
service_name = "nico nico seiga"
|
||||
author_name = sauce["data"]["member_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 9:
|
||||
service_name = "Danbooru"
|
||||
creator = sauce["data"]["creator"]
|
||||
material = sauce["data"]["material"]
|
||||
info = f"[{creator}]({material})"
|
||||
|
||||
elif sauce["header"]["index_id"] == 10:
|
||||
service_name = "drawr Images"
|
||||
author_name = sauce["data"]["member_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 11:
|
||||
service_name = "Nijie Images"
|
||||
author_name = sauce["data"]["member_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 12:
|
||||
service_name = "Yande.re"
|
||||
creator = sauce["data"]["creator"]
|
||||
material = sauce["data"]["material"]
|
||||
info = f"[{creator}]({material})"
|
||||
|
||||
# index 13 "animeop" disabled
|
||||
|
||||
# index 14 "IMDb" disabled
|
||||
|
||||
# index 15 "Shutterstock" disabled
|
||||
|
||||
elif sauce["header"]["index_id"] == 16:
|
||||
service_name = "FAKKU"
|
||||
creator = sauce["data"]["creator"]
|
||||
source = sauce["data"]["source"]
|
||||
info = f"[{creator}]({source})"
|
||||
|
||||
# index 17 reserved
|
||||
|
||||
elif sauce["header"]["index_id"] == 18 or sauce["header"]["index_id"] == 38:
|
||||
service_name = "H-Misc (ehentai)"
|
||||
eng_name = sauce["data"]["eng_name"]
|
||||
jp_name = sauce["data"]["jp_name"]
|
||||
info = f"{jp_name}" if jp_name else f"{eng_name}"
|
||||
|
||||
elif sauce["header"]["index_id"] == 19:
|
||||
service_name = "2D-Market"
|
||||
creator = sauce["data"]["creator"]
|
||||
source = sauce["data"]["source"]
|
||||
info = f"[{creator}]({source})"
|
||||
|
||||
elif sauce["header"]["index_id"] == 20:
|
||||
service_name = "MediBang"
|
||||
member_name = sauce["data"]["member_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{member_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 21:
|
||||
service_name = "Anime"
|
||||
title = sauce["data"]["source"]
|
||||
year = sauce["data"]["year"]
|
||||
part = sauce["data"]["part"]
|
||||
est_time = sauce["data"]["est_time"]
|
||||
time = est_time.split("/")[0]
|
||||
info = f"《{title}》/{year}\n第{part}集，{time}"
|
||||
|
||||
elif sauce["header"]["index_id"] == 22:
|
||||
service_name = "H-Anime"
|
||||
title = sauce["data"]["source"]
|
||||
year = sauce["data"]["year"]
|
||||
part = sauce["data"]["part"]
|
||||
est_time = sauce["data"]["est_time"]
|
||||
time = est_time.split("/")[0]
|
||||
info = f"《{title}》/{year}\n第{part}集，{time}"
|
||||
|
||||
elif sauce["header"]["index_id"] == 23:
|
||||
service_name = "IMDb-Movies"
|
||||
title = sauce["data"]["source"]
|
||||
year = sauce["data"]["year"]
|
||||
est_time = sauce["data"]["est_time"]
|
||||
time = est_time.split("/")[0]
|
||||
info = f"《{title}》/{year}，{time}"
|
||||
|
||||
elif sauce["header"]["index_id"] == 24:
|
||||
service_name = "IMDb-Shows"
|
||||
title = sauce["data"]["source"]
|
||||
year = sauce["data"]["year"]
|
||||
part = sauce["data"]["part"]
|
||||
est_time = sauce["data"]["est_time"]
|
||||
time = est_time.split("/")[0]
|
||||
info = f"《{title}》/{year}\n第{part}集，{time}"
|
||||
|
||||
elif sauce["header"]["index_id"] == 25:
|
||||
service_name = "Gelbooru"
|
||||
creator = sauce["data"]["creator"]
|
||||
material = sauce["data"]["material"]
|
||||
info = f"[{creator}]({material})"
|
||||
|
||||
elif sauce["header"]["index_id"] == 26:
|
||||
service_name = "Konachan"
|
||||
creator = sauce["data"]["creator"]
|
||||
material = sauce["data"]["material"]
|
||||
info = f"[{creator}]({material})"
|
||||
|
||||
elif sauce["header"]["index_id"] == 27:
|
||||
service_name = "Sankaku Channel"
|
||||
creator = sauce["data"]["creator"]
|
||||
material = sauce["data"]["material"]
|
||||
info = f"[{creator}]({material})"
|
||||
|
||||
elif sauce["header"]["index_id"] == 28:
|
||||
service_name = "Anime-Pictures.net"
|
||||
creator = sauce["data"]["creator"]
|
||||
material = sauce["data"]["material"]
|
||||
info = f"[{creator}]({material})"
|
||||
|
||||
elif sauce["header"]["index_id"] == 29:
|
||||
service_name = "e621.net"
|
||||
creator = sauce["data"]["creator"]
|
||||
material = sauce["data"]["material"]
|
||||
info = f"[{creator}]({material})"
|
||||
|
||||
elif sauce["header"]["index_id"] == 30:
|
||||
service_name = "Idol Complex"
|
||||
creator = sauce["data"]["creator"]
|
||||
material = sauce["data"]["material"]
|
||||
info = f"[{creator}]({material})"
|
||||
|
||||
elif sauce["header"]["index_id"] == 31:
|
||||
service_name = "bcy.net Illust"
|
||||
author_name = sauce["data"]["member_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 32:
|
||||
service_name = "bcy.net Cosplay"
|
||||
author_name = sauce["data"]["member_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 33:
|
||||
service_name = "PortalGraphics.net"
|
||||
member_name = sauce["data"]["member_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{member_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 34:
|
||||
service_name = "deviantArt"
|
||||
author_name = sauce["data"]["author_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 35:
|
||||
service_name = "Pawoo.net"
|
||||
illust_id = sauce["data"]["pawoo_id"]
|
||||
author_name = sauce["data"]["pawoo_user_display_name"]
|
||||
info = f"「{illust_id}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 36:
|
||||
service_name = "Madokami (Manga)"
|
||||
source = sauce["data"]["source"]
|
||||
part = sauce["data"]["part"]
|
||||
info = part if source in part else f"{source}-{part}"
|
||||
|
||||
elif sauce["header"]["index_id"] == 37 or sauce["header"]["index_id"] == 371:
|
||||
service_name = "MangaDex"
|
||||
artist = sauce["data"]["artist"]
|
||||
author = sauce["data"]["author"]
|
||||
source = sauce["data"]["source"]
|
||||
part = sauce["data"]["part"]
|
||||
info_a = f"[{artist}]" if artist == author else f"[{artist}({author})]"
|
||||
info_b = part if source in part else f"{source}-{part}"
|
||||
info = info_a + info_b
|
||||
|
||||
# index 38 "H-Misc (ehentai)" with 18
|
||||
|
||||
elif sauce["header"]["index_id"] == 39:
|
||||
service_name = "Artstation"
|
||||
author_name = sauce["data"]["author_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 40:
|
||||
service_name = "FurAffinity"
|
||||
author_name = sauce["data"]["author_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 41:
|
||||
service_name = "Twitter"
|
||||
author_name = sauce["data"]["twitter_user_handle"]
|
||||
time = sauce["data"]["created_at"]
|
||||
info = f"「{time[0:10]}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 42:
|
||||
service_name = "Furry Network"
|
||||
author_name = sauce["data"]["author_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「{author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 43:
|
||||
service_name = "Kemono"
|
||||
service = sauce["data"]["service_name"]
|
||||
author_name = sauce["data"]["user_name"]
|
||||
title = sauce["data"]["title"]
|
||||
info = f"「{title}」/「({service}){author_name}」"
|
||||
|
||||
elif sauce["header"]["index_id"] == 44:
|
||||
service_name = "Skeb"
|
||||
creator_name = sauce["data"]["creator_name"]
|
||||
creator = sauce["data"]["creator"]
|
||||
info = f"[{creator_name}]({creator})"
|
||||
|
||||
else:
|
||||
index = sauce["header"]["index_id"]
|
||||
service_name = f"Index #{index}"
|
||||
info = "no info"
|
||||
|
||||
except Exception as e:
|
||||
index = sauce["header"]["index_id"]
|
||||
service_name = f"Index #{index}"
|
||||
info = "no info"
|
||||
print(format_exc())
|
||||
|
||||
return service_name, info
|
||||
|
||||
|
||||
class SauceNAO:
|
||||
def __init__(
|
||||
self,
|
||||
api_key,
|
||||
output_type=2,
|
||||
testmode=0,
|
||||
dbmask=None,
|
||||
dbmaski=None,
|
||||
db=999,
|
||||
numres=3,
|
||||
shortlimit=20,
|
||||
longlimit=300,
|
||||
):
|
||||
params = dict()
|
||||
params["api_key"] = api_key
|
||||
params["output_type"] = output_type
|
||||
params["testmode"] = testmode
|
||||
params["dbmask"] = dbmask
|
||||
params["dbmaski"] = dbmaski
|
||||
params["db"] = db
|
||||
params["numres"] = numres
|
||||
self.params = params
|
||||
self.host = HOST_CUSTOM["SAUCENAO"] or "https://saucenao.com"
|
||||
self.header = "————>saucenao<————"
|
||||
|
||||
async def get_sauce(self, image_url):
|
||||
logger.debug(f"Now starting get the SauceNAO data:{image_url}")
|
||||
# 过滤 None 参数(SauceNAO 文件上传遇 dbmask=None 编码会返回 500)
|
||||
params = {k: v for k, v in self.params.items() if v is not None}
|
||||
# 优先本地下载图片后以文件形式提交,避免 SauceNAO 服务器抓取 QQ 图床失败
|
||||
try:
|
||||
async with _make_client(20) as client:
|
||||
resp = await client.get(image_url)
|
||||
resp.raise_for_status()
|
||||
image_bytes = resp.content
|
||||
submit_ok = True
|
||||
except Exception as e:
|
||||
logger.error(f"图片本地下载失败({e}), 退回 URL 提交")
|
||||
submit_ok = False
|
||||
|
||||
if submit_ok:
|
||||
params.pop("url", None)
|
||||
async with _make_client(30) as client:
|
||||
response = await client.post(
|
||||
f"{self.host}/search.php",
|
||||
params=params,
|
||||
files={"file": ("search_image.jpg", image_bytes, "image/jpeg")},
|
||||
)
|
||||
else:
|
||||
params["url"] = image_url
|
||||
async with _make_client(15) as client:
|
||||
response = await client.get(f"{self.host}/search.php", params=params)
|
||||
|
||||
if response.status_code != 200:
|
||||
raise RuntimeError(
|
||||
f"SauceNAO 请求失败: HTTP {response.status_code}, 响应: {response.text[:100]!r}"
|
||||
)
|
||||
return response.json()
|
||||
|
||||
async def get_view(self, sauce) -> str:
|
||||
sauces = await self.get_sauce(sauce)
|
||||
repass = ""
|
||||
simimax = 0
|
||||
index = 1
|
||||
|
||||
for sauce in sauces["results"]:
|
||||
try:
|
||||
url = (
|
||||
sauce["data"]["ext_urls"][0].replace("\\", "").strip()
|
||||
if "ext_urls" in sauce["data"]
|
||||
else "no link"
|
||||
)
|
||||
similarity = sauce["header"]["similarity"]
|
||||
if not similarity.replace(".", "").isdigit():
|
||||
similarity = 0
|
||||
simimax = float(similarity) if float(similarity) > simimax else simimax
|
||||
thumbnail_url = sauce["header"]["thumbnail"]
|
||||
if THUMB_ON:
|
||||
try:
|
||||
thumbnail_image = _img_cq(
|
||||
pic2b64(
|
||||
ats_pic(
|
||||
Image.open(BytesIO(await get_pic(thumbnail_url)))
|
||||
)
|
||||
)
|
||||
)
|
||||
except Exception as e:
|
||||
print(format_exc())
|
||||
thumbnail_image = "[预览图下载失败]"
|
||||
else:
|
||||
thumbnail_image = ""
|
||||
|
||||
service_name, info = sauces_info(sauce)
|
||||
|
||||
putline = (
|
||||
f"搜索结果{index}：\n\n{thumbnail_image}\n\n平台：\n\n{service_name}：\n\n"
|
||||
f"相似度:{similarity}%\n\n信息：\n\n{info}\n\n相关：\n\n{url}"
|
||||
)
|
||||
if repass:
|
||||
repass = "\n\n\n".join([repass, putline])
|
||||
else:
|
||||
repass = putline
|
||||
|
||||
except Exception as e:
|
||||
print(format_exc())
|
||||
pass
|
||||
index += 1
|
||||
return [repass, simimax]
|
||||
|
||||
|
||||
headers = {
|
||||
"Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9",
|
||||
"Accept-Encoding": "gzip, deflate",
|
||||
"Accept-Language": "zh-CN,zh;q=0.9",
|
||||
"Cache-Control": "max-age=0",
|
||||
"Connection": "keep-alive",
|
||||
"Origin": "https://ascii2d.net",
|
||||
"Referer": "https://ascii2d.net/",
|
||||
"Sec-Fetch-Dest": "document",
|
||||
"Sec-Fetch-Mode": "navigate",
|
||||
"Sec-Fetch-Site": "same-origin",
|
||||
"Sec-Fetch-User": "?1",
|
||||
"Upgrade-Insecure-Requests": "1",
|
||||
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.163 Safari/537.36",
|
||||
}
|
||||
|
||||
|
||||
class ascii2d:
|
||||
def __init__(self, num=2):
|
||||
self.num = num
|
||||
self.host = HOST_CUSTOM["ASCII"] or "https://ascii2d.net"
|
||||
self.header = "————>ascii2d<————"
|
||||
self.scraper = cloudscraper.create_scraper()
|
||||
|
||||
async def get_search_data(self, url: str, data=None):
|
||||
if data is not None:
|
||||
html = data
|
||||
else:
|
||||
html = await get_html_ascii2d(url)
|
||||
|
||||
all_data = html.xpath('//div[@class="row item-box"]')
|
||||
info = []
|
||||
for data in all_data[1 : self.num + 1]:
|
||||
try:
|
||||
title = ""
|
||||
member = ""
|
||||
if not data.xpath('.//img[@loading="lazy"]/@src'):
|
||||
continue
|
||||
thumb_url = data.xpath('.//img[@loading="lazy"]/@src')[0].strip()
|
||||
thumb_url = f"{self.host}{thumb_url}"
|
||||
|
||||
if not data.xpath('.//div[@class="detail-box gray-link"]/h6'):
|
||||
data2 = (
|
||||
data.xpath('.//div[@class="external"]')[0]
|
||||
if data.xpath('.//div[@class="external"]')
|
||||
else data
|
||||
)
|
||||
info_url = (
|
||||
data2.xpath(".//a/@href")[0].strip()
|
||||
if data2.xpath(".//a/@rel")
|
||||
else "no link"
|
||||
)
|
||||
tag = "外部登录" if info_url == "no link" else info_url.split("/")[2]
|
||||
else:
|
||||
data2 = data.xpath('.//div[@class="detail-box gray-link"]/h6')[0]
|
||||
info_url = data2.xpath(".//a/@href")[0].strip()
|
||||
tag = (
|
||||
data2.xpath("./small/text()") or data2.xpath(".//a/text()")
|
||||
)[0].strip()
|
||||
|
||||
if tag == "pixiv" or tag == "twitter":
|
||||
title = data2.xpath(".//a//text()")[0]
|
||||
member = data2.xpath(".//a//text()")[1]
|
||||
title = f"「{title}」/「{member}」"
|
||||
elif tag == "外部登录":
|
||||
title = data2.text.replace("\n", "") if data2.text else ""
|
||||
else:
|
||||
title = data2.text.replace("\n", "")
|
||||
title = f"「{title}」"
|
||||
|
||||
info.append([info_url, tag, thumb_url, title])
|
||||
except Exception as e:
|
||||
print(format_exc())
|
||||
logger.error(e)
|
||||
continue
|
||||
|
||||
return info
|
||||
|
||||
async def add_repass(self, tag: str, data, thumbnails: dict | None = None):
|
||||
po = "——{}——".format(tag)
|
||||
index = 1
|
||||
for line in data:
|
||||
if THUMB_ON:
|
||||
try:
|
||||
thumb_content = (thumbnails or {}).get(line[2])
|
||||
if thumb_content is None:
|
||||
thumb_content = self.scraper.get(
|
||||
line[2], timeout=20, proxies=proxies
|
||||
).content
|
||||
thumbnail_image = _img_cq(
|
||||
pic2b64(ats_pic(Image.open(BytesIO(thumb_content))))
|
||||
)
|
||||
except Exception as e:
|
||||
print(format_exc())
|
||||
thumbnail_image = "[预览图下载失败]"
|
||||
else:
|
||||
thumbnail_image = ""
|
||||
putline = (
|
||||
f"搜索结果{index}：\n\n{thumbnail_image}\n\n平台：\n\n{line[1]}\n\n"
|
||||
f"信息：\n\n{line[3]}\n\n相关：\n\n{line[0]}"
|
||||
)
|
||||
po = "\n\n\n".join([po, putline])
|
||||
index += 1
|
||||
|
||||
return po
|
||||
|
||||
async def get_view(self, ascii2d) -> str:
|
||||
putline1 = ""
|
||||
putline2 = ""
|
||||
url_index = f"{self.host}/search/url/{ascii2d}"
|
||||
logger.debug(f"Now starting get the {url_index}")
|
||||
|
||||
try:
|
||||
html_index = await get_html_ascii2d(url_index)
|
||||
if html_index is None:
|
||||
logger.error("ascii2d 页面获取失败(Cloudflare challenge 未通过)")
|
||||
return [putline1, putline2]
|
||||
except Exception as e:
|
||||
print(format_exc())
|
||||
logger.error(f"ascii2d get html data failed: {e}")
|
||||
return [putline1, putline2]
|
||||
|
||||
neet_div = html_index.xpath(
|
||||
'//div[@class="detail-link pull-xs-right hidden-sm-down gray-link"]'
|
||||
)
|
||||
|
||||
if neet_div:
|
||||
a_url_foot = neet_div[0].xpath("./span/a/@href")
|
||||
url2 = f"{self.host}{a_url_foot[1]}"
|
||||
|
||||
color = await self.get_search_data("", data=html_index)
|
||||
bovw = await self.get_search_data(url2)
|
||||
|
||||
# 用系统浏览器会话批量下载缩略图(带 CF cookie,避免被 Cloudflare 拦截)
|
||||
thumb_urls = [line[2] for line in color + bovw]
|
||||
thumbnails = await fetch_thumbs_playwright(thumb_urls)
|
||||
|
||||
if color:
|
||||
putline1 = await self.add_repass("色调检索", color, thumbnails)
|
||||
|
||||
if bovw:
|
||||
putline2 = await self.add_repass("特征检索", bovw, thumbnails)
|
||||
|
||||
return [putline1, putline2]
|
||||
|
||||
|
||||
async def get_image_data_sauce(image_url: str, api_key: str):
|
||||
if type(image_url) == list:
|
||||
image_url = image_url[0]
|
||||
|
||||
logger.info("Loading Image Search Container……")
|
||||
NAO = SauceNAO(api_key, numres=SAUCENAO_RESULT_NUM)
|
||||
|
||||
logger.debug("Loading all view……")
|
||||
repass = ""
|
||||
simimax = 0
|
||||
# 网络抖动/免费key限流时重试,退避间隔同时规避 SauceNAO 的 4 秒限流
|
||||
for attempt in range(3):
|
||||
try:
|
||||
result = await NAO.get_view(image_url)
|
||||
if result:
|
||||
header = NAO.header
|
||||
simimax = result[1]
|
||||
repass = "\n".join([header, result[0]])
|
||||
break
|
||||
except Exception as e:
|
||||
logger.error(f"SauceNAO 第{attempt + 1}次尝试失败: {e}")
|
||||
if attempt < 2:
|
||||
await asyncio.sleep(2 * (attempt + 1))
|
||||
else:
|
||||
return ["SauceNAO搜索失败……", 0]
|
||||
|
||||
return [repass, simimax]
|
||||
|
||||
|
||||
async def get_image_data_ascii(image_url: str):
|
||||
if type(image_url) == list:
|
||||
image_url = image_url[0]
|
||||
|
||||
logger.info("Loading Image Search Container……")
|
||||
ii2d = ascii2d(ASCII_RESULT_NUM)
|
||||
|
||||
logger.debug("Loading all view……")
|
||||
repass1 = ""
|
||||
repass2 = ""
|
||||
try:
|
||||
putline = await ii2d.get_view(image_url)
|
||||
if putline:
|
||||
header = ii2d.header
|
||||
if putline[0]:
|
||||
repass1 = "\n".join([header, putline[0]])
|
||||
if putline[1]:
|
||||
repass2 = "\n".join([header, putline[1]])
|
||||
except Exception as e:
|
||||
logger.error(format_exc())
|
||||
return ["ascii2d搜索失败……", ""]
|
||||
|
||||
return [repass1, repass2]
|
||||
|
||||
|
||||
# 系统浏览器可执行文件(与获取 cookie 的浏览器同款 TLS 指纹)
|
||||
CHROME_PATHS = [
|
||||
"C:/Program Files/Google/Chrome/Application/chrome.exe",
|
||||
"C:/Program Files (x86)/Google/Chrome/Application/chrome.exe",
|
||||
"C:/Program Files (x86)/Microsoft/Edge/Application/msedge.exe",
|
||||
"C:/Program Files/Microsoft/Edge/Application/msedge.exe",
|
||||
]
|
||||
|
||||
# 专用浏览器 profile 目录(cf_clearance 持久化,首次过验证后长期有效)
|
||||
CHROME_PROFILE_DIR = os.path.join(
|
||||
os.path.dirname(os.path.abspath(__file__)), "data", "chrome_profile"
|
||||
)
|
||||
|
||||
|
||||
def _parse_netscape_cookies(file_path: str) -> list[dict]:
|
||||
"""解析 Netscape 格式 cookies 文件(与 nonebot_plugin_video_analysis 同款)"""
|
||||
cookies = []
|
||||
with open(file_path, "r", encoding="utf-8") as f:
|
||||
for line in f:
|
||||
line = line.strip()
|
||||
if not line or line.startswith("#"):
|
||||
continue
|
||||
parts = line.split("\t")
|
||||
if len(parts) != 7:
|
||||
continue
|
||||
domain, flag, path, secure, expiry, name, value = parts
|
||||
cookie = {
|
||||
"name": name,
|
||||
"value": value,
|
||||
"domain": domain,
|
||||
"path": path,
|
||||
"secure": secure.upper() == "TRUE",
|
||||
"sameSite": "Lax",
|
||||
}
|
||||
if expiry.isdigit() and int(expiry) > 0:
|
||||
cookie["expires"] = int(expiry)
|
||||
cookies.append(cookie)
|
||||
return cookies
|
||||
|
||||
|
||||
def _load_ascii2d_cookies():
|
||||
"""读取配置的 ascii2d cookie 文件,未配置或不存在返回 None"""
|
||||
if not ASCII2D_COOKIES_FILE or not os.path.isfile(ASCII2D_COOKIES_FILE):
|
||||
return None
|
||||
return _parse_netscape_cookies(ASCII2D_COOKIES_FILE)
|
||||
|
||||
|
||||
def _save_netscape_cookies(cookies: list[dict], file_path: str):
|
||||
"""把浏览器回传的 cookies 写回 Netscape 格式文件(实现 cf_clearance 自愈)"""
|
||||
lines = ["# Netscape HTTP Cookie File"]
|
||||
for c in cookies:
|
||||
domain = c.get("domain", "")
|
||||
include_subdomains = "TRUE" if domain.startswith(".") else "FALSE"
|
||||
path = c.get("path", "/")
|
||||
secure = "TRUE" if c.get("secure") else "FALSE"
|
||||
try:
|
||||
expires = int(c.get("expires") or 0)
|
||||
except (TypeError, ValueError):
|
||||
expires = 0
|
||||
name = c.get("name", "")
|
||||
value = c.get("value", "")
|
||||
lines.append(
|
||||
f"{domain}\t{include_subdomains}\t{path}\t{secure}\t{expires}\t{name}\t{value}"
|
||||
)
|
||||
with open(file_path, "w", encoding="utf-8") as f:
|
||||
f.write("\n".join(lines))
|
||||
|
||||
|
||||
async def get_html_ascii2d(url):
|
||||
"""获取 ascii2d 页面 HTML:配置了 cookie 文件时注入系统浏览器直过验证"""
|
||||
return await get_html_content(url, cookies=_load_ascii2d_cookies())
|
||||
|
||||
|
||||
async def fetch_thumbs_playwright(urls: list[str]) -> dict:
|
||||
"""用系统浏览器会话批量下载缩略图(带 CF cookie,避免 Cloudflare 拦截)"""
|
||||
if not urls:
|
||||
return {}
|
||||
result = {}
|
||||
async with async_playwright() as p:
|
||||
launch_kwargs = {
|
||||
"headless": False,
|
||||
"args": [
|
||||
"--window-position=-32000,-32000",
|
||||
"--disable-blink-features=AutomationControlled",
|
||||
],
|
||||
}
|
||||
for path in CHROME_PATHS:
|
||||
if os.path.isfile(path):
|
||||
launch_kwargs["executable_path"] = path
|
||||
break
|
||||
context = await p.chromium.launch_persistent_context(
|
||||
CHROME_PROFILE_DIR, **launch_kwargs
|
||||
)
|
||||
cookies = _load_ascii2d_cookies()
|
||||
if cookies:
|
||||
await context.add_cookies(cookies)
|
||||
try:
|
||||
for u in urls:
|
||||
try:
|
||||
resp = await context.request.get(u, timeout=20000)
|
||||
if resp.status == 200 and "image" in resp.headers.get(
|
||||
"content-type", ""
|
||||
):
|
||||
result[u] = await resp.body()
|
||||
except Exception as e:
|
||||
logger.debug(f"缩略图下载失败 {u}: {e}")
|
||||
finally:
|
||||
await context.close()
|
||||
return result
|
||||
|
||||
|
||||
async def get_html_content(url, cookies=None):
|
||||
async with async_playwright() as p:
|
||||
# 用系统 Chrome + 固定 profile 目录(cf_clearance 持久化,与浏览器获取 cookie 时同款 TLS 指纹)
|
||||
# 不指定 UA:使用系统 Chrome 真实 UA,否则 cf_clearance(与 UA 绑定)会失效
|
||||
# headless 会被 Cloudflare 检测拒绝,用有头模式并将窗口移出屏幕避免打扰
|
||||
launch_kwargs = {
|
||||
"headless": False,
|
||||
"args": [
|
||||
"--window-position=-32000,-32000",
|
||||
"--disable-blink-features=AutomationControlled",
|
||||
],
|
||||
}
|
||||
for path in CHROME_PATHS:
|
||||
if os.path.isfile(path):
|
||||
launch_kwargs["executable_path"] = path
|
||||
break
|
||||
context = await p.chromium.launch_persistent_context(
|
||||
CHROME_PROFILE_DIR, **launch_kwargs
|
||||
)
|
||||
await context.add_init_script(
|
||||
"Object.defineProperty(navigator, 'webdriver', {get: () => undefined});"
|
||||
)
|
||||
if cookies:
|
||||
await context.add_cookies(cookies)
|
||||
|
||||
page = context.pages[0] if context.pages else await context.new_page()
|
||||
|
||||
try:
|
||||
await page.goto(url, wait_until="domcontentloaded")
|
||||
# 轮询等待 Cloudflare challenge 自动通过(最多约 20 秒)
|
||||
passed = False
|
||||
for _ in range(8):
|
||||
await page.wait_for_timeout(2500)
|
||||
title = await page.title()
|
||||
if "Just a moment" not in title:
|
||||
passed = True
|
||||
break
|
||||
html = etree.HTML(await page.content())
|
||||
# 浏览器可能刷新了 cf_clearance,回写文件实现自愈
|
||||
if passed and ASCII2D_COOKIES_FILE:
|
||||
try:
|
||||
_save_netscape_cookies(
|
||||
await context.cookies(), ASCII2D_COOKIES_FILE
|
||||
)
|
||||
logger.info("ascii2d cookies 已刷新回写")
|
||||
except Exception as e:
|
||||
logger.error(f"ascii2d cookies 回写失败: {e}")
|
||||
print(f"获取到的html内容：{html}")
|
||||
return html
|
||||
except Exception as e:
|
||||
print(f"Error fetching page: {e}")
|
||||
return None
|
||||
finally:
|
||||
await context.close()
|
||||
Reference in New Issue
Block a user