✨ 支持发病小作文查重、随机生成

defs/asoulcnki.py

@@ -0,0 +1,114 @@
+import time
+from io import BytesIO
+from PIL import Image
+import httpx
+import jinja2
+import random
+from os import sep
+from init import logger
+from defs.browser import html_to_pic
+from defs.diff import diff_text
+
+env = jinja2.Environment(enable_async=True)
+with open(f"resources{sep}templates{sep}article.html", "r", encoding="utf-8") as f:
+    article_data = f.read()
+article_tpl = env.from_string(article_data)
+
+
+async def check_text(text: str):
+    try:
+        url = 'https://asoulcnki.asia/v1/api/check'
+        async with httpx.AsyncClient() as client:
+            resp = await client.post(url=url, json={'text': text})
+            result = resp.json()
+
+        if result['code'] != 0:
+            return None, None
+
+        data = result['data']
+        if not data['related']:
+            return None, '没有找到重复的小作文捏'
+
+        rate = data['rate']
+        related = data['related'][0]
+        reply_url = str(related['reply_url']).strip()
+        reply = related['reply']
+
+        msg = ['枝网文本复制检测报告',
+               '',
+               '总复制比 {:.2f}%'.format(rate * 100),
+               f'相似小作文: <a href="{reply_url}">地点</a> - '
+               f'{time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(reply["ctime"]))}',]
+
+        image = await render_reply(reply, diff=text)
+        if not image:
+            return None, "\n".join(msg)
+        return image, "\n".join(msg)
+    except Exception as e:
+        logger.warning(f"Error in check_text: {e}")
+        return None, None
+
+
+async def random_text(keyword: str = ""):
+    try:
+        url = 'https://asoulcnki.asia/v1/api/ranking'
+        params = {
+            'pageSize': 10,
+            'pageNum': 1,
+            'timeRangeMode': 0,
+            'sortMode': 0
+        }
+        if keyword:
+            params['keywords'] = keyword
+        else:
+            params['pageNum'] = random.randint(1, 100)
+
+        async with httpx.AsyncClient() as client:
+            resp = await client.get(url=url, params=params)
+            result = resp.json()
+
+        if result['code'] != 0:
+            return None, None
+
+        replies = result['data']['replies']
+        if not replies:
+            return None, '没有找到小作文捏'
+
+        reply = random.choice(replies)
+        image = await render_reply(reply)
+        reply_url = f"https://t.bilibili.com/{reply['dynamic_id']}/#reply{reply['rpid']}"
+        if not image:
+            return None, f'<a href="{reply_url}">转到小作文</a>'
+        return image, f'<a href="{reply_url}">转到小作文</a>'
+    except Exception as e:
+        logger.warning(f"Error in random_text: {e}")
+        return None, None
+
+
+async def render_reply(reply: dict, diff: str = ""):
+    try:
+        article = {}
+        article['username'] = reply['m_name']
+        article['like'] = reply['like_num']
+        article['all_like'] = reply['similar_like_sum']
+        article['quote'] = reply['similar_count']
+        article['text'] = diff_text(
+            diff, reply['content']) if diff else reply['content']
+        article['time'] = time.strftime(
+            "%Y-%m-%d", time.localtime(reply['ctime']))
+
+        html = await article_tpl.render_async(article=article)
+        img_raw = await html_to_pic(html, wait=0, viewport={"width": 500, "height": 100})
+        # 将bytes结果转化为字节流
+        bytes_stream = BytesIO(img_raw)
+        # 读取到图片
+        img = Image.open(bytes_stream)
+        imgByteArr = BytesIO()  # 初始化一个空字节流
+        img.save(imgByteArr, format('PNG'))  # 把我们得图片以 PNG 保存到空字节流
+        imgByteArr = imgByteArr.getvalue()  # 无视指针，获取全部内容，类型由io流变成bytes。
+        with open(f"data{sep}asoulcnki.png", 'wb') as i:
+            i.write(imgByteArr)
+        return f"data{sep}asoulcnki.png"
+    except Exception as e:
+        logger.warning(f"Error in render_reply: {e}")
+        return None

defs/browser.py

@@ -0,0 +1,108 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+@Author         : yanyongyu
+@Date           : 2021-03-12 13:42:43
+@LastEditors    : yanyongyu
+@LastEditTime   : 2021-11-01 14:05:41
+@Description    : None
+@GitHub         : https://github.com/yanyongyu
+"""
+__author__ = "yanyongyu"
+
+import asyncio
+import platform
+from contextlib import asynccontextmanager
+from os import getcwd
+from typing import Optional, AsyncIterator
+from playwright.async_api import Page, Browser, async_playwright, Error
+from init import logger
+from uvicorn.loops import asyncio as _asyncio
+from uvicorn import config
+
+
+def asyncio_setup():
+    loop = asyncio.new_event_loop()
+    asyncio.set_event_loop(loop)
+
+
+@property
+def should_reload(self):
+    return False
+
+
+if platform.system() == "Windows":
+    _asyncio.asyncio_setup = asyncio_setup
+    config.Config.should_reload = should_reload
+    logger.warning("检测到当前为 Windows 系统，已自动注入猴子补丁")
+
+_browser: Optional[Browser] = None
+_playwright = None
+
+
+async def init(**kwargs) -> Browser:
+    global _browser
+    global _playwright
+    _playwright = await async_playwright().start()
+    try:
+        _browser = await launch_browser(**kwargs)
+    except Error:
+        await install_browser()
+        _browser = await launch_browser(**kwargs)
+    return _browser
+
+
+async def launch_browser(**kwargs) -> Browser:
+    return await _playwright.chromium.launch(**kwargs)
+
+
+async def get_browser(**kwargs) -> Browser:
+    return _browser or await init(**kwargs)
+
+
+@asynccontextmanager
+async def get_new_page(**kwargs) -> AsyncIterator[Page]:
+    browser = await get_browser()
+    page = await browser.new_page(**kwargs)
+    try:
+        yield page
+    finally:
+        await page.close()
+
+
+async def shutdown_browser():
+    await _browser.close()
+    await _playwright.stop()
+
+
+async def install_browser():
+    logger.info("正在安装 chromium")
+    import sys
+    from playwright.__main__ import main
+    sys.argv = ['', 'install', 'chromium']
+    try:
+        main()
+    except SystemExit:
+        pass
+
+
+async def html_to_pic(
+    html: str, wait: int = 0, template_path: str = f"file://{getcwd()}", **kwargs
+) -> bytes:
+    """html转图片
+    Args:
+        html (str): html文本
+        wait (int, optional): 等待时间. Defaults to 0.
+        template_path (str, optional): 模板路径 如 "file:///path/to/template/"
+    Returns:
+        bytes: 图片, 可直接发送
+    """
+    # logger.debug(f"html:\n{html}")
+    if "file:" not in template_path:
+        raise "template_path 应该为 file:///path/to/template"
+    async with get_new_page(**kwargs) as page:
+        await page.goto(template_path)
+        await page.set_content(html, wait_until="networkidle")
+        await page.wait_for_timeout(wait)
+        img_raw = await page.screenshot(full_page=True)
+    return img_raw

defs/diff.py

@@ -0,0 +1,100 @@
+from typing import List
+
+
+class cache:
+    def __init__(self, start, end):
+        self.start = start
+        self.end = end
+
+    def __eq__(self, other):
+        if isinstance(other, self.__class__):
+            return self.start == other.start and self.end == other.end
+        else:
+            return False
+
+
+def merge(intervals: List[cache]):
+    """合并重复区间
+    :param intervals: 待去重区间
+    :returns: 去重的区间（根据开始位置逆序）
+    """
+    if len(intervals) == 0:
+        return []
+    intervals = sorted(intervals, key=lambda s: s.start)
+    outputs = [intervals[0]]
+    for s in intervals:
+        last_interval = outputs[-1]
+        if last_interval.end < s.start:
+            outputs.append(s)
+        else:
+            last_interval.end = max(last_interval.end, s.end)
+    return sorted(outputs, key=lambda s: s.start, reverse=True)
+
+
+def compare(origin: str, dest: str, sensitive: int):
+    """标记重复区间
+    :param origin: 待查重文本
+    :param dest: 返回的文本
+    :param sensitive: 敏感长度
+    :returns: 重复区间数组（根据开始位置逆序）
+    """
+    length = max(len(origin), len(dest)) ** 2
+    matrix = [0 for i in range(length)]
+    cache_array: List[cache] = []
+
+    def convert(index_y: int, index_x: int):
+        return index_y * len(origin) + index_x
+
+    def remove(arr: List[cache], obj: cache):
+        return list(filter(lambda s: s != obj, arr))
+
+    def new_cache(end: int, offset: int):
+        start = end - offset
+        start = 0 if start < 0 else start + 1
+        return cache(start, offset + start)
+
+    for index, s in enumerate(origin):
+        if dest[0] == s:
+            matrix[index] = 1
+
+    for index_x, x in enumerate(dest):
+        for index_y, y in enumerate(origin):
+            index = convert(index_y, index_x)
+            pre_index = convert(index_y - 1, index_x - 1)
+            if x == y:
+                if index_y == 0:
+                    matrix[index] = 1
+                    continue
+                matrix[index] = matrix[pre_index] + 1
+                if matrix[index] >= sensitive:
+                    cache_array.append(new_cache(index_y, matrix[index]))
+                if matrix[index] > sensitive:
+                    cache_array = remove(cache_array, new_cache(
+                        index_y - 1, matrix[pre_index]))
+    return merge(cache_array)
+
+
+def render(s: str, flag: List[cache], tag: str):
+    """给重复区间加tag
+    :param s: raw text
+    :param flag: repeat area Array
+    :param tag: used tag, default em
+    :returns: tagged text
+    """
+    arr = list(s)
+    for i in flag:
+        arr.insert(i.end, f'</{tag}>')
+        arr.insert(i.start, f'<{tag}>')
+    return ''.join(arr)
+
+
+def diff_text(origin: str, dest: str, sensitive=4, tag='strong'):
+    """对文本重复对比，给重复部分加tag
+    :param origin: 待查重文本
+    :param dest: 服务器返回的文本
+    :param sensitive: 敏感长度
+    :param tag: HTML tag, example a, em
+    :returns: 做好标记的文本
+    """
+    flag = compare(dest, origin, sensitive)
+    return render(dest, flag, tag)

init.py

@@ -1,7 +1,4 @@
 import logging
-
-from pyrogram.types import User
-
 from defs.glover import ipv6
 from pyrogram import Client
 from logging import getLogger, INFO, ERROR, StreamHandler, basicConfig

modules/ask.py

@@ -10,6 +10,8 @@ from defs.ask import how_many, what_time, how_long, hif, handle_pers, who
                    filters.regex(r"^问"))
 async def ask(client: Client, message: Message):
     msg = message
+    if not message.text:
+        raise ContinuePropagation
     message = message.text.strip()[1:]
     handled = False
     if re.findall("几|多少", message):

modules/asoulcnki.py

@@ -0,0 +1,45 @@
+from pyrogram import Client, filters, ContinuePropagation
+from pyrogram.types import Message
+from defs.asoulcnki import check_text, random_text
+from init import user_me
+
+
+@Client.on_message(filters.incoming & ~filters.edited &
+                   filters.command(["查重", f"查重@{user_me.username}"]))
+async def asoulcnki_process(client: Client, message: Message):
+    text = message.reply_to_message.text if message.reply_to_message else " ".join(message.text.split(" ")[1:])
+    if not text:
+        await message.reply("请输入文本。")
+        raise ContinuePropagation
+    if len(text) >= 1000:
+        await message.reply('文本过长，长度须在10-1000之间', quote=True)
+        raise ContinuePropagation
+    elif len(text) <= 10:
+        await message.reply('文本过短，长度须在10-1000之间', quote=True)
+        raise ContinuePropagation
+    image, text = await check_text(text)
+    if image:
+        await message.reply_photo(image, quote=True, caption=text)
+    else:
+        if text:
+            await message.reply(text, quote=True)
+        else:
+            await message.reply('出错了，请稍后再试', quote=True)
+    raise ContinuePropagation
+
+
+@Client.on_message(filters.incoming & ~filters.edited &
+                   filters.command(["小作文", f"小作文@{user_me.username}"]))
+async def asoulcnki_random(client: Client, message: Message):
+    text = message.reply_to_message.text if message.reply_to_message else " ".join(message.text.split(" ")[1:])
+    if not text:
+        text = ""
+    image, text = await random_text(text)
+    if image:
+        await message.reply_photo(image, quote=True, caption=text)
+    else:
+        if text:
+            await message.reply(text, quote=True)
+        else:
+            await message.reply('出错了，请稍后再试', quote=True)
+    raise ContinuePropagation

modules/start.py

@@ -11,6 +11,7 @@ des = """本机器人特性：
 ★ 我有个朋友
 ★ 简易版问与答
 ★ 网易云音乐
+★ 查重、小作文
 """
 
 

requirements.txt

@@ -7,3 +7,6 @@ coloredlogs
 qrcode
 pyncm
 mutagen
+playwright
+uvicorn
+jinja2

resources/templates/article.html

@@ -0,0 +1,107 @@
+<html>
+  <head>
+    <style>
+      @font-face {
+        font-family: "iconfont";
+        src: url("https://at.alicdn.com/t/font_2949364_zcqb63am41o.woff2?t=1637832961945")
+          format("woff2");
+      }
+      .iconfont {
+        font-family: "iconfont" !important;
+        font-size: 16px;
+        font-style: normal;
+        -webkit-font-smoothing: antialiased;
+        -moz-osx-font-smoothing: grayscale;
+      }
+      .icon-like:before {
+        content: "\e6bf";
+      }
+      .icon-all-like:before {
+        content: "\e6c1";
+      }
+      .icon-qoute:before {
+        content: "\e620";
+      }
+      .icon-user:before {
+        content: "\e6f4";
+      }
+      .icon-time:before {
+        content: "\e703";
+      }
+      body {
+        background-color: rgba(243, 244, 246, 1);
+      }
+      .article {
+        display: flex;
+        flex-direction: column;
+        border-radius: 5px;
+        border-color: rgba(229, 231, 235, 1);
+        box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1);
+        background-color: rgba(249, 250, 251, 1);
+      }
+      .article-header {
+        display: flex;
+        flex-direction: row;
+        align-items: center;
+        margin-left: 8px;
+        margin-right: 8px;
+        padding: 8px;
+      }
+      .article-footer {
+        display: flex;
+        flex-direction: row;
+        justify-content: space-between;
+        align-items: center;
+        margin-left: 8px;
+        margin-right: 8px;
+        padding: 8px;
+      }
+      .article-text {
+        margin-left: 8px;
+        margin-right: 8px;
+        padding: 8px;
+        border-radius: 5px;
+        background-color: rgba(229, 231, 235, 1);
+        line-height: 1.625;
+        word-break: break-all;
+        white-space: pre-wrap;
+      }
+      .article-text strong {
+        font-weight: 400;
+        background-color: rgba(243, 244, 246, 1);
+        color: rgba(239, 68, 68, 1);
+      }
+    </style>
+  </head>
+  <body>
+    <div class="image">
+      <div class="article">
+        <div class="article-header">
+          <div>
+            <span class="iconfont icon-user"></span>
+            {{ article["username"] }}
+          </div>
+        </div>
+        <div class="article-text">{{ article["text"] }}</div>
+        <div class="article-footer">
+          <div>
+            <span class="iconfont icon-all-like" />
+            {{ article["all_like"] }}
+          </div>
+          <div>
+            <span class="iconfont icon-like" />
+            {{ article["like"] }}
+          </div>
+          <div>
+            <span class="iconfont icon-qoute" />
+            {{ article["quote"] }}
+          </div>
+          <div>
+            <span class="iconfont icon-time" />
+            {{ article["time"] }}
+          </div>
+        </div>
+      </div>
+    </div>
+  </body>
+</html>