610 lines
19 KiB
Python
610 lines
19 KiB
Python
"""Markdown 转换器 - 支持 HTML / PNG / Plain Text 输出
|
||
|
||
Dependencies:
|
||
- HTML/PNG 输出: pip install Pillow html2text pygments markdown
|
||
"""
|
||
|
||
import html
|
||
import re
|
||
import sys
|
||
import os
|
||
from pathlib import Path
|
||
|
||
|
||
# ============================================================
|
||
# 字体查找
|
||
# ============================================================
|
||
|
||
def _find_font():
|
||
"""Find a suitable TrueType font across platforms."""
|
||
candidates = []
|
||
if sys.platform == "win32":
|
||
pf = os.environ.get("WINDIR", r"C:\Windows")
|
||
candidates = [
|
||
os.path.join(pf, "Fonts", "msyh.ttc"),
|
||
os.path.join(pf, "Fonts", "msyhbd.ttc"),
|
||
os.path.join(pf, "Fonts", "simhei.ttf"),
|
||
os.path.join(pf, "Fonts", "simsun.ttc"),
|
||
]
|
||
elif sys.platform == "darwin":
|
||
candidates = [
|
||
"/System/Library/Fonts/PingFang.ttc",
|
||
"/System/Library/Fonts/STHeiti Light.ttc",
|
||
"/Library/Fonts/Arial Unicode.ttf",
|
||
]
|
||
else:
|
||
candidates = [
|
||
"/usr/share/fonts/truetype/noto/NotoSansCJK-Regular.ttc",
|
||
"/usr/share/fonts/opentype/noto/NotoSansCJK-Regular.ttc",
|
||
"/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf",
|
||
"/usr/share/fonts/truetype/wqy/wqy-microhei.ttc",
|
||
]
|
||
for p in candidates:
|
||
if os.path.exists(p):
|
||
return p
|
||
return None
|
||
|
||
|
||
# ============================================================
|
||
# 工具函数
|
||
# ============================================================
|
||
|
||
def _clean_invisible_chars(text):
|
||
"""清理不可见的 Unicode 字符"""
|
||
if not text:
|
||
return ""
|
||
text = re.sub(r'[\u200b-\u200f\u2028-\u202f\ufeff\u00ad]', '', text)
|
||
text = re.sub(r'[\xa0\u3000]', ' ', text)
|
||
text = re.sub(r'[\uff00-\uffef]', '', text)
|
||
return text
|
||
|
||
|
||
def _decode_html_entities(text):
|
||
"""解码 HTML 实体"""
|
||
entities = {
|
||
'<': '<',
|
||
'>': '>',
|
||
'&': '&',
|
||
'"': '"',
|
||
''': "'",
|
||
''': "'",
|
||
}
|
||
for k, v in entities.items():
|
||
text = text.replace(k, v)
|
||
# 处理特殊引号
|
||
text = text.replace('"', '"').replace('"', '"')
|
||
text = text.replace(''', "'").replace(''', "'")
|
||
return text
|
||
|
||
|
||
# ============================================================
|
||
# Markdown 解析
|
||
# ============================================================
|
||
|
||
def parse_markdown(md_text):
|
||
"""解析 Markdown 文本，提取标题、代码块、段落等元素"""
|
||
if not md_text:
|
||
return []
|
||
|
||
md_text = _clean_invisible_chars(md_text)
|
||
lines = md_text.split('\n')
|
||
elements = []
|
||
current_paragraph = []
|
||
|
||
def flush_paragraph():
|
||
nonlocal current_paragraph
|
||
if current_paragraph:
|
||
text = ' '.join(current_paragraph)
|
||
if text.strip():
|
||
elements.append(('paragraph', text.strip()))
|
||
current_paragraph = []
|
||
|
||
for line in lines:
|
||
stripped = line.strip()
|
||
|
||
# 跳过空行
|
||
if not stripped:
|
||
flush_paragraph()
|
||
continue
|
||
|
||
# 标题
|
||
header_match = re.match(r'^(#{1,6})\s+(.+)$', stripped)
|
||
if header_match:
|
||
flush_paragraph()
|
||
level = len(header_match.group(1))
|
||
elements.append(('header', level, header_match.group(2).strip()))
|
||
continue
|
||
|
||
# 代码块
|
||
if stripped.startswith('```'):
|
||
flush_paragraph()
|
||
elements.append(('codeblock', stripped[3:].strip()))
|
||
continue
|
||
|
||
# 无序列表
|
||
list_match = re.match(r'^[\-\*+]\s+(.+)$', stripped)
|
||
if list_match:
|
||
flush_paragraph()
|
||
elements.append(('list_item', list_match.group(1).strip()))
|
||
continue
|
||
|
||
# 有序列表
|
||
ordered_match = re.match(r'^\d+\.\s+(.+)$', stripped)
|
||
if ordered_match:
|
||
flush_paragraph()
|
||
elements.append(('ordered_item', ordered_match.group(1).strip()))
|
||
continue
|
||
|
||
# 引用
|
||
if stripped.startswith('>'):
|
||
flush_paragraph()
|
||
content = stripped[1:].strip()
|
||
elements.append(('quote', content))
|
||
continue
|
||
|
||
# 水平线
|
||
if re.match(r'^[\-\*_]{3,}$', stripped):
|
||
flush_paragraph()
|
||
elements.append(('hr',))
|
||
continue
|
||
|
||
# 链接或图片
|
||
link_match = re.match(r'!?\[([^\]]+)\]\([^\)]+\)', stripped)
|
||
if link_match:
|
||
flush_paragraph()
|
||
elements.append(('link', link_match.group(1)))
|
||
continue
|
||
|
||
# 默认作为段落处理
|
||
current_paragraph.append(stripped)
|
||
|
||
flush_paragraph()
|
||
return elements
|
||
|
||
|
||
# ============================================================
|
||
# 文本换行
|
||
# ============================================================
|
||
|
||
def _wrap_text(text, font, max_width, draw):
|
||
"""Wrap text to fit within max_width pixels."""
|
||
lines = []
|
||
for paragraph in text.split("\n"):
|
||
if not paragraph.strip():
|
||
lines.append("")
|
||
continue
|
||
current = ""
|
||
for ch in paragraph:
|
||
test = current + ch
|
||
bbox = draw.textbbox((0, 0), test, font=font)
|
||
if bbox[2] - bbox[0] > max_width and current:
|
||
lines.append(current)
|
||
current = ch
|
||
else:
|
||
current = test
|
||
if current:
|
||
lines.append(current)
|
||
return lines
|
||
|
||
|
||
# ============================================================
|
||
# HTML 转换
|
||
# ============================================================
|
||
|
||
def markdown_to_html(md_text):
|
||
"""将 Markdown 转换为 HTML"""
|
||
if not md_text:
|
||
return ""
|
||
|
||
md_text = _clean_invisible_chars(md_text)
|
||
elements = parse_markdown(md_text)
|
||
|
||
try:
|
||
import html
|
||
except ImportError:
|
||
import urllib.parse as html
|
||
|
||
try:
|
||
import markdown
|
||
from markdown.extensions import codehilite
|
||
|
||
md = markdown.Markdown(extensions=['codehilite', 'fenced_code', 'tables'])
|
||
html_content = md.convert(md_text)
|
||
return _html_template(html_content)
|
||
except ImportError:
|
||
# 降级处理：使用简单的转换
|
||
return _simple_markdown_to_html(md_text)
|
||
|
||
|
||
def _simple_markdown_to_html(md_text):
|
||
"""简单的 Markdown 到 HTML 转换（无外部依赖）"""
|
||
lines = md_text.split('\n')
|
||
html_lines = []
|
||
|
||
in_codeblock = False
|
||
|
||
for line in lines:
|
||
stripped = line.strip()
|
||
|
||
# 代码块开始/结束
|
||
if stripped.startswith('```'):
|
||
if in_codeblock:
|
||
html_lines.append('</code></pre>')
|
||
in_codeblock = False
|
||
else:
|
||
lang = stripped[3:].strip() or ''
|
||
lang_attr = f' class="language-{lang}"' if lang else ''
|
||
html_lines.append(f'<pre><code lang="{lang}">')
|
||
in_codeblock = True
|
||
continue
|
||
|
||
if in_codeblock:
|
||
html_lines.append(html.escape(line))
|
||
continue
|
||
|
||
# 标题
|
||
header_match = re.match(r'^(#{1,6})\s+(.+)$', stripped)
|
||
if header_match:
|
||
level = len(header_match.group(1))
|
||
content = header_match.group(2)
|
||
html_lines.append(f'<h{level}>{_decode_html_entities(content)}</h{level}>')
|
||
continue
|
||
|
||
# 水平线
|
||
if re.match(r'^[\-\*_]{3,}$', stripped):
|
||
html_lines.append('<hr>')
|
||
continue
|
||
|
||
# 引用
|
||
if stripped.startswith('>'):
|
||
content = stripped[1:].strip()
|
||
html_lines.append(f'<blockquote>{_decode_html_entities(content)}</blockquote>')
|
||
continue
|
||
|
||
# 列表项
|
||
list_match = re.match(r'^[\-\*+]\s+(.+)$', stripped)
|
||
if list_match:
|
||
content = list_match.group(1)
|
||
html_lines.append(f'<li>{_decode_html_entities(content)}</li>')
|
||
continue
|
||
|
||
# 段落
|
||
if stripped:
|
||
# 处理粗体和斜体
|
||
text = _decode_html_entities(stripped)
|
||
text = re.sub(r'\*\*(.+?)\*\*', r'<strong>\1</strong>', text)
|
||
text = re.sub(r'\*(.+?)\*', r'<em>\1</em>', text)
|
||
text = re.sub(r'__(.+?)__', r'<strong>\1</strong>', text)
|
||
text = re.sub(r'_(.+?)_', r'<em>\1</em>', text)
|
||
text = re.sub(r'`(.+?)`', r'<code>\1</code>', text)
|
||
html_lines.append(f'<p>{text}</p>')
|
||
|
||
return _html_template('\n'.join(html_lines))
|
||
|
||
|
||
def _html_template(content):
|
||
"""生成完整的 HTML 文档"""
|
||
return f"""<!DOCTYPE html>
|
||
<html lang="zh-CN">
|
||
<head>
|
||
<meta charset="UTF-8">
|
||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||
<title>Markdown Document</title>
|
||
<style>
|
||
body {{
|
||
font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, 'Helvetica Neue', Arial, sans-serif;
|
||
line-height: 1.6;
|
||
max-width: 800px;
|
||
margin: 0 auto;
|
||
padding: 20px;
|
||
color: #333;
|
||
}}
|
||
h1, h2, h3, h4, h5, h6 {{
|
||
margin-top: 1.5em;
|
||
margin-bottom: 0.5em;
|
||
font-weight: 600;
|
||
}}
|
||
h1 {{ font-size: 2em; border-bottom: 2px solid #333; }}
|
||
h2 {{ font-size: 1.5em; border-bottom: 1px solid #ddd; }}
|
||
code {{
|
||
background: #f4f4f4;
|
||
padding: 2px 6px;
|
||
border-radius: 3px;
|
||
font-family: 'Consolas', 'Monaco', monospace;
|
||
}}
|
||
pre {{
|
||
background: #f4f4f4;
|
||
padding: 16px;
|
||
border-radius: 6px;
|
||
overflow-x: auto;
|
||
}}
|
||
pre code {{
|
||
background: none;
|
||
padding: 0;
|
||
}}
|
||
blockquote {{
|
||
border-left: 4px solid #ddd;
|
||
margin: 0;
|
||
padding-left: 16px;
|
||
color: #666;
|
||
}}
|
||
a {{
|
||
color: #0066cc;
|
||
}}
|
||
hr {{
|
||
border: none;
|
||
border-top: 1px solid #ddd;
|
||
margin: 24px 0;
|
||
}}
|
||
</style>
|
||
</head>
|
||
<body>
|
||
{content}
|
||
</body>
|
||
</html>"""
|
||
|
||
|
||
# ============================================================
|
||
# PNG 渲染 (使用 matplotlib)
|
||
# ============================================================
|
||
|
||
def _get_matplotlib_font():
|
||
"""获取支持中文的 matplotlib 字体（通过字体文件路径）"""
|
||
import matplotlib
|
||
import matplotlib.font_manager as fm
|
||
import os
|
||
import sys
|
||
|
||
# Linux 中文字体路径
|
||
linux_font_paths = [
|
||
'/usr/share/fonts/opentype/noto/NotoSansCJK-Regular.ttc',
|
||
'/usr/share/fonts/opentype/noto/NotoSansSC-Regular.otf',
|
||
'/usr/share/fonts/truetype/noto/NotoSansCJK-Regular.ttc',
|
||
'/usr/share/fonts/truetype/wqy/wqy-microhei.ttc',
|
||
'/usr/share/fonts/truetype/wqy/wqy-microhei.ttc',
|
||
'/usr/share/fonts/truetype/droid/DroidSansFallbackFull.ttf',
|
||
'/usr/share/fonts/google-noto-cjk/NotoSansCJK-Regular.ttc',
|
||
'/usr/share/fonts/truetype/arphic/uming.ttc',
|
||
'/usr/share/fonts/truetype/arphic/ukai.ttc',
|
||
]
|
||
|
||
# Windows 中文字体路径
|
||
if sys.platform == "win32":
|
||
windir = os.environ.get("WINDIR", r"C:\Windows")
|
||
windows_font_paths = [
|
||
os.path.join(windir, "Fonts", "msyh.ttc"),
|
||
os.path.join(windir, "Fonts", "msyhbd.ttc"),
|
||
os.path.join(windir, "Fonts", "simhei.ttf"),
|
||
os.path.join(windir, "Fonts", "simsun.ttc"),
|
||
os.path.join(windir, "Fonts", "STHeiti Light.ttc"),
|
||
]
|
||
linux_font_paths.extend(windows_font_paths)
|
||
|
||
# macOS 中文字体路径
|
||
elif sys.platform == "darwin":
|
||
mac_font_paths = [
|
||
'/System/Library/Fonts/PingFang.ttc',
|
||
'/System/Library/Fonts/STHeiti Light.ttc',
|
||
'/Library/Fonts/Arial Unicode.ttf',
|
||
'/System/Library/Fonts/Supplemental/Arial Unicode.ttf',
|
||
]
|
||
linux_font_paths.extend(mac_font_paths)
|
||
|
||
# 查找存在的字体文件
|
||
for font_path in linux_font_paths:
|
||
if os.path.exists(font_path):
|
||
# 清除字体缓存并加载指定字体
|
||
fm.fontManager.addfont(font_path)
|
||
font = fm.FontProperties(fname=font_path)
|
||
# 验证字体可以显示中文
|
||
return font
|
||
|
||
# 如果没找到，返回 None 让 matplotlib 使用默认
|
||
return None
|
||
|
||
|
||
def markdown_to_png(md_text, img_path):
|
||
"""将 Markdown 渲染为 PNG 图片（使用 matplotlib）"""
|
||
try:
|
||
import matplotlib.pyplot as plt
|
||
import matplotlib.patches as patches
|
||
import matplotlib.font_manager as fm
|
||
import matplotlib
|
||
except ImportError:
|
||
raise ImportError("matplotlib not installed. Run: pip install matplotlib")
|
||
|
||
# 设置非交互式后端
|
||
matplotlib.use('Agg')
|
||
|
||
# 获取中文字体
|
||
font = _get_matplotlib_font()
|
||
|
||
elements = parse_markdown(md_text)
|
||
|
||
W, PAD = 10, 0.5 # 英寸, 边距
|
||
FIG_H = 2.0 # 初始高度
|
||
LINE_H = 0.35 # 每行高度
|
||
CODE_H = 0.5 # 代码块初始高度
|
||
|
||
# 计算所需高度
|
||
y = 2.5 # 顶部空间
|
||
for elem in elements:
|
||
if elem[0] == 'header':
|
||
level = elem[1]
|
||
text = elem[2]
|
||
chars_per_line = 50 if level <= 2 else 60
|
||
lines = max(1, len(text) // chars_per_line + 1)
|
||
y += lines * (0.5 if level <= 2 else 0.4) + 0.2
|
||
elif elem[0] == 'paragraph':
|
||
chars_per_line = 60
|
||
lines = max(1, len(elem[1]) // chars_per_line + 1)
|
||
y += lines * 0.35 + 0.3
|
||
elif elem[0] == 'codeblock':
|
||
lines = elem[1].count('\n') + 2
|
||
y += lines * 0.3 + 0.2
|
||
elif elem[0] in ('list_item', 'quote'):
|
||
chars_per_line = 55
|
||
lines = max(1, len(elem[1]) // chars_per_line + 1)
|
||
y += lines * 0.32 + 0.15
|
||
|
||
FIG_H = max(8, y)
|
||
|
||
fig, ax = plt.subplots(figsize=(W, FIG_H))
|
||
fig.patch.set_facecolor('#ffffff')
|
||
ax.set_facecolor('#ffffff')
|
||
|
||
# 标题栏背景
|
||
header = patches.Rectangle((0, FIG_H - 1.2), W, 1.0, linewidth=0, facecolor='#2d3748')
|
||
ax.add_patch(header)
|
||
|
||
# 标题
|
||
ax.text(0.5, FIG_H - 0.6, 'Markdown Document',
|
||
fontsize=18, fontweight='bold', color='white',
|
||
fontproperties=font, ha='left', va='center')
|
||
|
||
ax.text(0.5, FIG_H - 1.0, 'Converted from Markdown',
|
||
fontsize=10, color='#888888', fontproperties=font, ha='left', va='center')
|
||
|
||
ax.set_xlim(0, W)
|
||
ax.set_ylim(0, FIG_H)
|
||
ax.axis('off')
|
||
|
||
cy = FIG_H - 1.5
|
||
|
||
for elem in elements:
|
||
if elem[0] == 'header':
|
||
level = elem[1]
|
||
text = elem[2]
|
||
size = 16 if level <= 2 else 14
|
||
weight = 'bold' if level == 1 else 'normal'
|
||
color = '#1a1a2e' if level == 1 else '#2d3748'
|
||
ax.text(PAD, cy, text, fontsize=size, fontweight=weight,
|
||
color=color, fontproperties=font, ha='left', va='top')
|
||
cy -= size * 0.04 + 0.15
|
||
|
||
elif elem[0] == 'paragraph':
|
||
ax.text(PAD, cy, elem[1], fontsize=11, color='#374151',
|
||
fontproperties=font, ha='left', va='top', wrap=True)
|
||
lines = max(1, len(elem[1]) // 60 + 1)
|
||
cy -= lines * 0.35 + 0.25
|
||
|
||
elif elem[0] == 'codeblock':
|
||
code_h = max(0.5, (elem[1].count('\n') + 2) * 0.3)
|
||
code_box = patches.Rectangle((PAD, cy - code_h), W - PAD * 2, code_h,
|
||
linewidth=1, edgecolor='#e0e0e0', facecolor='#f4f4f4')
|
||
ax.add_patch(code_box)
|
||
ax.text(PAD + 0.1, cy - 0.15, elem[1][:500], fontsize=9,
|
||
color='#333333', fontfamily='monospace', va='top')
|
||
cy -= code_h + 0.2
|
||
|
||
elif elem[0] == 'list_item':
|
||
ax.text(PAD, cy, f'\u2022 {elem[1]}', fontsize=11, color='#374151',
|
||
fontproperties=font, ha='left', va='top')
|
||
cy -= 0.35
|
||
|
||
elif elem[0] == 'quote':
|
||
ax.plot([PAD, PAD, PAD + 0.05, PAD + 0.05],
|
||
[cy, cy - 0.4, cy - 0.4, cy - 0.6],
|
||
color='#0066cc', linewidth=2)
|
||
ax.text(PAD + 0.15, cy - 0.1, elem[1], fontsize=11, color='#666666',
|
||
fontproperties=font, ha='left', va='top')
|
||
cy -= 0.5
|
||
|
||
elif elem[0] == 'hr':
|
||
ax.axhline(y=cy, color='#e0e0e0', linewidth=1, xmin=0.05, xmax=0.95)
|
||
cy -= 0.3
|
||
|
||
plt.tight_layout(pad=0)
|
||
|
||
Path(img_path).parent.mkdir(parents=True, exist_ok=True)
|
||
plt.savefig(img_path, format='png', dpi=150, bbox_inches='tight',
|
||
facecolor='#ffffff', edgecolor='none')
|
||
plt.close()
|
||
|
||
|
||
# ============================================================
|
||
# 纯文本转换
|
||
# ============================================================
|
||
|
||
def markdown_to_text(md_text):
|
||
"""将 Markdown 转换为纯文本"""
|
||
if not md_text:
|
||
return ""
|
||
|
||
md_text = _clean_invisible_chars(md_text)
|
||
elements = parse_markdown(md_text)
|
||
|
||
lines = []
|
||
|
||
for elem in elements:
|
||
if elem[0] == 'header':
|
||
level = elem[1]
|
||
prefix = "#" * level + " "
|
||
lines.append(f"{prefix}{elem[2]}")
|
||
lines.append("")
|
||
|
||
elif elem[0] == 'paragraph':
|
||
lines.append(elem[1])
|
||
lines.append("")
|
||
|
||
elif elem[0] == 'codeblock':
|
||
lines.append("```")
|
||
lines.append(elem[1])
|
||
lines.append("```")
|
||
lines.append("")
|
||
|
||
elif elem[0] == 'list_item':
|
||
lines.append(f"• {elem[1]}")
|
||
|
||
elif elem[0] == 'ordered_item':
|
||
lines.append(f" {elem[1]}")
|
||
|
||
elif elem[0] == 'quote':
|
||
lines.append(f"> {elem[1]}")
|
||
|
||
elif elem[0] == 'hr':
|
||
lines.append("─" * 50)
|
||
lines.append("")
|
||
|
||
return '\n'.join(lines)
|
||
|
||
|
||
# ============================================================
|
||
# 主函数
|
||
# ============================================================
|
||
|
||
def main():
|
||
if len(sys.argv) < 3:
|
||
print("Usage: python md_convert.py <input.md> <output.{html|png|txt}>", file=sys.stderr)
|
||
sys.exit(1)
|
||
|
||
input_path = sys.argv[1]
|
||
output_path = sys.argv[2]
|
||
|
||
if not os.path.exists(input_path):
|
||
print(f"Error: Input file not found: {input_path}", file=sys.stderr)
|
||
sys.exit(1)
|
||
|
||
md_text = Path(input_path).read_text(encoding="utf-8")
|
||
ext = Path(output_path).suffix.lower()
|
||
|
||
Path(output_path).parent.mkdir(parents=True, exist_ok=True)
|
||
|
||
if ext == ".html":
|
||
html_content = markdown_to_html(md_text)
|
||
Path(output_path).write_text(html_content, encoding="utf-8")
|
||
elif ext == ".png":
|
||
markdown_to_png(md_text, output_path)
|
||
elif ext == ".txt":
|
||
text_content = markdown_to_text(md_text)
|
||
Path(output_path).write_text(text_content, encoding="utf-8")
|
||
else:
|
||
print(f"Error: Unsupported output format: {ext}", file=sys.stderr)
|
||
print("Supported formats: .html, .png, .txt", file=sys.stderr)
|
||
sys.exit(1)
|
||
|
||
print(f"Converted: {input_path} -> {output_path}")
|
||
|
||
|
||
if __name__ == "__main__":
|
||
main() |