Mr.Xia
1b484d7fda
- damo_balance/__init__.py: 将外层字符串改为单引号，消除内嵌双引号引起的 SyntaxError - chatai/screenshot.py: bleach 改为 try/except 可选导入，无 bleach 时降级跳过 HTML 净化 - requirements.txt: 补充 openai>=1.0.0 与 pyppeteer>=1.0.2 依赖声明
196 lines
7.3 KiB
Python
196 lines
7.3 KiB
Python
import asyncio
|
||
import re
|
||
import html as html_module
|
||
import markdown
|
||
try:
|
||
import bleach
|
||
_HAS_BLEACH = True
|
||
except ImportError:
|
||
_HAS_BLEACH = False
|
||
from nonebot import logger
|
||
|
||
async def markdown_to_image(markdown_text: str, output_path: str, browser=None):
|
||
"""将 Markdown 转换为 HTML 并使用 Puppeteer 截图。"""
|
||
page = None
|
||
should_close_browser = False
|
||
try:
|
||
# Convert markdown to HTML. The markdown library handles special chars safely.
|
||
# Note: do NOT html.escape() before markdown.markdown() - it breaks markdown syntax.
|
||
html_content = markdown.markdown(markdown_text, extensions=["fenced_code", "tables"])
|
||
# Sanitize to prevent XSS from malicious AI responses
|
||
allowed_tags = [
|
||
'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p', 'br', 'hr',
|
||
'ul', 'ol', 'li', 'blockquote', 'pre', 'code', 'span',
|
||
'table', 'thead', 'tbody', 'tr', 'th', 'td',
|
||
'strong', 'em', 'b', 'i', 'u', 'a', 'img', 'div',
|
||
]
|
||
allowed_attrs = {'a': ['href', 'title'], 'img': ['src', 'alt', 'title']}
|
||
if _HAS_BLEACH:
|
||
html_content = bleach.clean(html_content, tags=allowed_tags, attributes=allowed_attrs)
|
||
else:
|
||
logger.warning("[chatai] bleach 未安装，跳过 HTML 净化（请运行 pip install bleach）")
|
||
|
||
# 使用传入的浏览器实例或创建新的
|
||
if browser is None:
|
||
from pyppeteer import launch
|
||
browser = await launch(headless=True, args=['--no-sandbox', '--disable-setuid-sandbox'])
|
||
should_close_browser = True
|
||
|
||
page = await browser.newPage()
|
||
page.setDefaultNavigationTimeout(15000)
|
||
|
||
# 设置页面样式，使内容更美观
|
||
await page.setContent(f"""
|
||
<html>
|
||
<head>
|
||
<style>
|
||
body {{
|
||
font-family: "PingFang SC", "Microsoft YaHei", sans-serif;
|
||
line-height: 1.6;
|
||
padding: 30px;
|
||
max-width: 800px;
|
||
margin: 0 auto;
|
||
background-color: transparent;
|
||
color: #333;
|
||
}}
|
||
.container {{
|
||
background-color: #ffffff;
|
||
border-radius: 15px;
|
||
box-shadow: 0 4px 12px rgba(0, 0, 0, 0.1);
|
||
padding: 25px;
|
||
overflow: hidden;
|
||
}}
|
||
p {{
|
||
margin-bottom: 16px;
|
||
}}
|
||
code {{
|
||
font-family: "SFMono-Regular", Consolas, "Liberation Mono", Menlo, monospace;
|
||
background-color: #f5f7f9;
|
||
padding: 2px 6px;
|
||
border-radius: 4px;
|
||
font-size: 0.9em;
|
||
}}
|
||
pre {{
|
||
background-color: #f5f7f9;
|
||
padding: 15px;
|
||
border-radius: 8px;
|
||
overflow-x: auto;
|
||
margin: 20px 0;
|
||
}}
|
||
pre code {{
|
||
background-color: transparent;
|
||
padding: 0;
|
||
}}
|
||
h1, h2, h3, h4, h5, h6 {{
|
||
margin-top: 24px;
|
||
margin-bottom: 16px;
|
||
font-weight: 600;
|
||
line-height: 1.25;
|
||
}}
|
||
h1 {{
|
||
font-size: 1.8em;
|
||
border-bottom: 1px solid #eaecef;
|
||
padding-bottom: 0.3em;
|
||
}}
|
||
h2 {{
|
||
font-size: 1.5em;
|
||
border-bottom: 1px solid #eaecef;
|
||
padding-bottom: 0.3em;
|
||
}}
|
||
blockquote {{
|
||
padding: 0 1em;
|
||
color: #6a737d;
|
||
border-left: 0.25em solid #dfe2e5;
|
||
margin: 16px 0;
|
||
}}
|
||
ul, ol {{
|
||
padding-left: 2em;
|
||
margin-bottom: 16px;
|
||
}}
|
||
img {{
|
||
max-width: 100%;
|
||
border-radius: 8px;
|
||
}}
|
||
a {{
|
||
color: #0366d6;
|
||
text-decoration: none;
|
||
}}
|
||
a:hover {{
|
||
text-decoration: underline;
|
||
}}
|
||
table {{
|
||
border-collapse: collapse;
|
||
width: 100%;
|
||
margin: 16px 0;
|
||
}}
|
||
table th, table td {{
|
||
padding: 8px 12px;
|
||
border: 1px solid #dfe2e5;
|
||
}}
|
||
table th {{
|
||
background-color: #f6f8fa;
|
||
}}
|
||
</style>
|
||
</head>
|
||
<body>
|
||
<div class="container">
|
||
{html_content}
|
||
</div>
|
||
</body>
|
||
</html>
|
||
""")
|
||
|
||
# 等待内容渲染完成
|
||
try:
|
||
await asyncio.wait_for(page.waitForNavigation({'waitUntil': 'networkidle0'}), timeout=10)
|
||
except Exception:
|
||
pass # rendering may already be complete
|
||
|
||
# 获取内容尺寸并设置视口
|
||
dimensions = await page.evaluate('''() => {
|
||
const container = document.querySelector('.container');
|
||
return {
|
||
width: container.offsetWidth + 60, // 加上 body 的 padding
|
||
height: container.offsetHeight + 60
|
||
}
|
||
}''')
|
||
|
||
# 设置视口大小
|
||
await page.setViewport({
|
||
'width': dimensions['width'],
|
||
'height': dimensions['height'],
|
||
'deviceScaleFactor': 2 # 提高图片清晰度
|
||
})
|
||
|
||
# 截图，使用透明背景
|
||
await page.screenshot({
|
||
'path': output_path,
|
||
'omitBackground': True, # 透明背景
|
||
'clip': {
|
||
'x': 0,
|
||
'y': 0,
|
||
'width': dimensions['width'],
|
||
'height': dimensions['height']
|
||
}
|
||
})
|
||
|
||
# 关闭页面
|
||
await page.close()
|
||
|
||
# 如果是我们创建的浏览器实例，则关闭它
|
||
if should_close_browser:
|
||
await browser.close()
|
||
|
||
except Exception as e:
|
||
# 确保资源被释放
|
||
if page is not None:
|
||
try:
|
||
await page.close()
|
||
except Exception:
|
||
pass
|
||
if should_close_browser and browser is not None:
|
||
try:
|
||
await browser.close()
|
||
except Exception:
|
||
pass
|
||
raise # 重新抛出异常以便上层处理 |