md2wx-cli 0.2.0__py3-none-any.whl

md2wx/__init__.py

@@ -0,0 +1,3 @@
+"""md2wx — Markdown to WeChat HTML CLI tool."""
+
+__version__ = "0.2.0"

md2wx/cli.py

@@ -0,0 +1,299 @@
+"""md2wx CLI 入口点 —— Markdown to WeChat HTML 排版与生产力套件."""
+
+import argparse
+import sys
+from pathlib import Path
+from typing import List, Optional
+
+from . import __version__
+from .clipboard import copy_html_to_clipboard
+from .config import get_wechat_credentials
+from .converter import WeChatConverter, make_paste_safe, preview_html
+from .fetcher import fetch_and_convert_article
+from .gallery import launch_gallery
+from .publisher import publish_to_wechat
+from .theme import Theme, list_themes, load_theme
+from .theme_extractor import extract_from_dir, extract_from_file, extract_from_url, save_theme_yaml
+from .validator import print_diagnostic_report, validate_wechat_article
+
+
+def _load_effective_theme(theme_name: str, theme_file: Optional[str]) -> Theme:
+    if theme_file:
+        import yaml
+
+        path = Path(theme_file)
+        if not path.exists():
+            raise FileNotFoundError(f"未找到外部主题文件: {theme_file}")
+        data = yaml.safe_load(path.read_text(encoding="utf-8-sig"))
+        if not isinstance(data, dict):
+            raise ValueError(f"外部主题 YAML 格式无效 (必须为字典映射): {theme_file}")
+        theme = Theme(
+            name=data.get("name", path.stem),
+            description=data.get("description", ""),
+            base_css=data.get("base_css", ""),
+            colors=data.get("colors", {}) if isinstance(data.get("colors"), dict) else {},
+            _raw_data=data,
+        )
+        return theme
+    return load_theme(theme_name)
+
+
+def cmd_convert(args: argparse.Namespace) -> int:
+    in_path = None
+    if args.input == "-" or (not args.input and not sys.stdin.isatty()):
+        input_text = sys.stdin.read()
+        input_filename = "document"
+    elif args.input:
+        in_path = Path(args.input)
+        if not in_path.exists():
+            print(f"错误: 输入文件未找到: {args.input}", file=sys.stderr)
+            return 1
+        input_text = in_path.read_text(encoding="utf-8-sig")
+        input_filename = in_path.stem
+    else:
+        print("错误: 请指定输入的 Markdown 文件或使用子命令 (运行 'md2wx --help' 查看帮助)", file=sys.stderr)
+        return 1
+
+    try:
+        theme = _load_effective_theme(args.theme, args.theme_file)
+    except Exception as e:
+        print(f"加载主题失败: {e}", file=sys.stderr)
+        return 1
+
+    converter = WeChatConverter(theme=theme)
+    base_dir = in_path.parent if in_path else None
+    result = converter.convert(input_text, base_dir=base_dir)
+    body_html = result.html
+
+    if not getattr(args, "no_paste_safe", False):
+        body_html = make_paste_safe(body_html)
+
+    final_output = preview_html(body_html, theme) if getattr(args, "full_page", False) else body_html
+
+    out_file = args.output
+    if out_file:
+        out_path = Path(out_file)
+        out_path.parent.mkdir(parents=True, exist_ok=True)
+        out_path.write_text(final_output, encoding="utf-8")
+        print(f"已生成文件: {out_path}", file=sys.stderr)
+
+    if getattr(args, "clipboard", False):
+        plain = input_text
+        ok = copy_html_to_clipboard(body_html, plain)
+        if ok:
+            print("已成功复制富文本 HTML 到系统剪贴板！可直接去微信公众号后台 Ctrl+V 粘贴。", file=sys.stderr)
+        else:
+            print("写入剪贴板失败，请使用 -o 保存为文件后手动复制。", file=sys.stderr)
+
+    if getattr(args, "publish", False):
+        creds = get_wechat_credentials(
+            cli_appid=getattr(args, "appid", None),
+            cli_secret=getattr(args, "secret", None),
+            cli_author=getattr(args, "author", None),
+        )
+        if not creds.appid or not creds.secret:
+            print(
+                "错误: 发布至微信草稿箱需要提供公众号凭据 (AppID 与 AppSecret)。\n"
+                "请通过命令行参数 --appid/--secret，或配置环境变量 WECHAT_APPID 与 WECHAT_SECRET，"
+                "或在 ~/.md2wx/config.yaml 中配置。",
+                file=sys.stderr,
+            )
+            return 1
+
+        title = getattr(args, "title", "") or result.title or input_filename
+        digest = getattr(args, "digest", "") or result.digest
+        cover = getattr(args, "cover", None)
+
+        print(f"正在发布到微信公众号草稿箱: 《{title}》...")
+        try:
+            media_id = publish_to_wechat(
+                html=body_html,
+                title=title,
+                digest=digest,
+                images=result.images,
+                appid=creds.appid,
+                secret=creds.secret,
+                cover=cover,
+                author=creds.author,
+                base_dir=base_dir,
+            )
+            print(f"[OK] 成功发布至微信公众号草稿箱！media_id: {media_id}")
+            print("可登录 https://mp.weixin.qq.com 后台在「草稿箱」中查看并群发。")
+            return 0
+        except Exception as e:
+            print(f"[ERROR] 发布失败: {e}", file=sys.stderr)
+            return 1
+
+    if not out_file and not getattr(args, "clipboard", False) and not getattr(args, "publish", False):
+        print(final_output)
+
+    return 0
+
+
+def cmd_gallery(argv: List[str]) -> int:
+    parser = argparse.ArgumentParser(prog="md2wx gallery", description="生成 19 套主题的 Shadcn 交互式画廊并在浏览器中打开")
+    parser.add_argument("input", help="输入的 Markdown 文件路径")
+    parser.add_argument("-o", "--output", help="输出的画廊 HTML 路径（可选）")
+    args = parser.parse_args(argv)
+
+    p = Path(args.input)
+    if not p.exists():
+        print(f"错误: 输入文件未找到: {args.input}", file=sys.stderr)
+        return 1
+
+    out_p = Path(args.output) if args.output else None
+    target = launch_gallery(p, output_html=out_p)
+    print(f"[OK] 已成功生成主题画廊并已在默认浏览器中打开: {target}")
+    return 0
+
+
+def cmd_learn_theme(argv: List[str]) -> int:
+    parser = argparse.ArgumentParser(
+        prog="md2wx learn-theme",
+        description="从微信在线文章或本地设计目录 (如 test_html2json) 逆向学习并生成新主题",
+    )
+    group = parser.add_mutually_exclusive_group(required=True)
+    group.add_argument("--url", help="微信公众号文章在线 URL")
+    group.add_argument("--dir", help="本地设计快照目录（包含 DESIGN.md 或 HTML 资源）")
+    group.add_argument("--file", help="单个本地 HTML 或 DESIGN.md 文件路径")
+    parser.add_argument("-n", "--name", required=True, help="要生成的主题名称（如 my-brand）")
+    parser.add_argument("-o", "--output", help="保存主题的目标文件路径（默认存入全局 ~/.md2wx/themes/）")
+    args = parser.parse_args(argv)
+
+    try:
+        if args.url:
+            yaml_str = extract_from_url(args.url, theme_name=args.name)
+        elif args.dir:
+            yaml_str = extract_from_dir(Path(args.dir), theme_name=args.name)
+        elif args.file:
+            yaml_str = extract_from_file(Path(args.file), theme_name=args.name)
+        else:
+            print("错误: 请指定 --url, --dir 或 --file", file=sys.stderr)
+            return 1
+
+        out_path = Path(args.output) if args.output else None
+        target_path = save_theme_yaml(yaml_str, args.name, target_path=out_path)
+        print(f"[OK] 成功提取并生成新主题: {args.name}")
+        print(f"[PATH] 已保存至: {target_path}")
+        print(f"[TIP] 立即使用: md2wx article.md -t {args.name} -c")
+        return 0
+    except Exception as e:
+        print(f"提取主题失败: {e}", file=sys.stderr)
+        return 1
+
+
+def cmd_fetch(argv: List[str]) -> int:
+    parser = argparse.ArgumentParser(prog="md2wx fetch", description="抓取微信公众号文章并逆向提取为本地 Markdown与图片")
+    parser.add_argument("url", help="微信推文 URL")
+    parser.add_argument("-o", "--output", help="输出的 Markdown 文件路径（默认按标题生成）")
+    args = parser.parse_args(argv)
+
+    try:
+        out_p = Path(args.output) if args.output else None
+        target_md, _ = fetch_and_convert_article(args.url, output_file=out_p)
+        print(f"[OK] 微信推文已成功抓取并转为 Markdown: {target_md}")
+        return 0
+    except Exception as e:
+        print(f"抓取文章失败: {e}", file=sys.stderr)
+        return 1
+
+
+def cmd_check(argv: List[str]) -> int:
+    parser = argparse.ArgumentParser(prog="md2wx check", description="对 Markdown 或 HTML 进行微信排版兼容性与合规体检")
+    parser.add_argument("file", help="要检查的 Markdown 或 HTML 文件路径")
+    parser.add_argument("-t", "--theme", default="default", help="排版主题名称 (默认: default)")
+    args = parser.parse_args(argv)
+
+    p = Path(args.file)
+    if not p.exists():
+        print(f"错误: 文件未找到: {args.file}", file=sys.stderr)
+        return 1
+
+    diagnostics = validate_wechat_article(p, theme_name=args.theme)
+    print_diagnostic_report(p, diagnostics)
+    return 1 if any(d.level == "ERROR" for d in diagnostics) else 0
+
+
+def cmd_themes_list(argv: List[str]) -> int:
+    names = list_themes()
+    print(f"已安装的排版主题列表 (共 {len(names)} 个):")
+    for n in names:
+        try:
+            t = load_theme(n)
+            print(f"  - {n:<20} {t.description}")
+        except Exception:
+            print(f"  - {n:<20}")
+    return 0
+
+
+def build_parser() -> argparse.ArgumentParser:
+    epilog_text = """常用功能与子命令:
+  md2wx article.md -c                          # 转换并一键复制富文本到剪贴板
+  md2wx article.md -t impeccable -p            # 原生直传公众号草稿箱（自动换取微信 CDN 图床）
+  md2wx gallery article.md                     # 生成 19 套主题的 Shadcn 交互式对比画廊
+  md2wx learn-theme --dir test_html2json -n my # 从本地设计快照提取新主题
+  md2wx learn-theme --url <url> -n tech        # 从微信在线推文提取新主题
+  md2wx fetch <url> -o my-article.md           # 抓取微信推文转为 Markdown 与本地图片
+  md2wx check article.md                       # 对文章执行微信排版兼容性诊断体检
+  md2wx themes                                 # 查看全部已安装主题
+"""
+    parser = argparse.ArgumentParser(
+        prog="md2wx",
+        description="Markdown to WeChat HTML CLI tool (支持画廊对比、主题提取、公众号草稿直传、推文抓取、排版体检)",
+        epilog=epilog_text,
+        formatter_class=argparse.RawDescriptionHelpFormatter,
+    )
+    parser.add_argument("-v", "-V", "--version", action="version", version=f"md2wx {__version__}", help="显示版本号并退出")
+    parser.add_argument("input", nargs="?", help="输入的 Markdown 文件路径（传入 - 或省略时支持管道标准输入）")
+    parser.add_argument("-o", "--output", help="输出的 HTML 文件路径")
+    parser.add_argument("-t", "--theme", default="default", help="排版主题名称 (默认: default)")
+    parser.add_argument("--theme-file", help="外部自定义 YAML 主题文件路径")
+    parser.add_argument("-c", "--clipboard", action="store_true", help="转换后直接将富文本 HTML 写入系统剪贴板")
+    parser.add_argument("--full-page", action="store_true", help="生成包含完整 <html> 结构的独立预览页面")
+    parser.add_argument("-p", "--publish", action="store_true", help="原生直传并发布至微信公众号草稿箱（需 AppID/Secret）")
+    parser.add_argument("--appid", help="微信公众号 AppID")
+    parser.add_argument("--secret", help="微信公众号 AppSecret")
+    parser.add_argument("--cover", help="图文封面图片路径")
+    parser.add_argument("--title", help="文章标题（覆盖 Markdown 一级标题）")
+    parser.add_argument("--author", help="文章作者名称")
+    parser.add_argument("--digest", help="文章摘要简介")
+    parser.add_argument("--no-paste-safe", action="store_true", help="关闭 span[leaf] 粘贴防护")
+    return parser
+
+
+def main(argv: Optional[List[str]] = None) -> int:
+    if argv is None:
+        argv = sys.argv[1:]
+
+    if not argv:
+        parser = build_parser()
+        parser.print_help()
+        return 0
+
+    cmd = argv[0].lower()
+
+    if cmd in ("gallery", "preview-all"):
+        return cmd_gallery(argv[1:])
+    elif cmd in ("learn-theme", "extract-theme"):
+        return cmd_learn_theme(argv[1:])
+    elif cmd == "fetch":
+        return cmd_fetch(argv[1:])
+    elif cmd in ("check", "validate", "doctor"):
+        return cmd_check(argv[1:])
+    elif cmd in ("themes", "list-themes"):
+        return cmd_themes_list(argv[1:])
+    elif cmd in ("version", "--version", "-v", "-V"):
+        print(f"md2wx {__version__}")
+        return 0
+    elif cmd in ("help", "--help", "-h"):
+        parser = build_parser()
+        parser.print_help()
+        return 0
+
+    parser = build_parser()
+    args = parser.parse_args(argv)
+    return cmd_convert(args)
+
+
+if __name__ == "__main__":
+    sys.exit(main())

md2wx/clipboard.py

@@ -0,0 +1,169 @@
+"""跨平台剪贴板处理模块 —— 支持富文本 HTML 与纯文本写入。"""
+
+import os
+import platform
+import subprocess
+import sys
+from typing import Optional
+
+
+def _format_windows_cf_html(html_fragment: str) -> bytes:
+    """构建 Windows 标准的 CF_HTML 数据包格式。"""
+    header_template = (
+        "Version:0.9\r\n"
+        "StartHTML:{start_html:08d}\r\n"
+        "EndHTML:{end_html:08d}\r\n"
+        "StartFragment:{start_fragment:08d}\r\n"
+        "EndFragment:{end_fragment:08d}\r\n"
+    )
+
+    prefix = (
+        '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">\r\n'
+        '<HTML><HEAD><META http-equiv="content-type" content="text/html; charset=utf-8"></HEAD><BODY>\r\n'
+        "<!--StartFragment-->"
+    )
+    suffix = "<!--EndFragment--></BODY></HTML>"
+
+    dummy_header = header_template.format(start_html=0, end_html=0, start_fragment=0, end_fragment=0)
+    header_len = len(dummy_header.encode("utf-8"))
+    prefix_len = len(prefix.encode("utf-8"))
+    body_len = len(html_fragment.encode("utf-8"))
+    suffix_len = len(suffix.encode("utf-8"))
+
+    start_html = header_len
+    start_fragment = header_len + prefix_len
+    end_fragment = start_fragment + body_len
+    end_html = end_fragment + suffix_len
+
+    header = header_template.format(
+        start_html=start_html,
+        end_html=end_html,
+        start_fragment=start_fragment,
+        end_fragment=end_fragment,
+    )
+    return (header + prefix + html_fragment + suffix).encode("utf-8")
+
+
+def _copy_windows(html_str: str, plain_text: str) -> bool:
+    """Windows 原生 Win32 API 写入 CF_HTML 与 CF_UNICODETEXT。"""
+    try:
+        import ctypes
+        from ctypes import wintypes
+
+        user32 = ctypes.windll.user32
+        kernel32 = ctypes.windll.kernel32
+
+        # 必须显式声明 64 位指针类型，避免 64 位内存地址被截断为 32 位 int
+        kernel32.GlobalAlloc.restype = wintypes.HGLOBAL
+        kernel32.GlobalAlloc.argtypes = [wintypes.UINT, ctypes.c_size_t]
+
+        kernel32.GlobalLock.restype = wintypes.LPVOID
+        kernel32.GlobalLock.argtypes = [wintypes.HGLOBAL]
+
+        kernel32.GlobalUnlock.restype = wintypes.BOOL
+        kernel32.GlobalUnlock.argtypes = [wintypes.HGLOBAL]
+
+        user32.RegisterClipboardFormatW.restype = wintypes.UINT
+        user32.RegisterClipboardFormatW.argtypes = [wintypes.LPCWSTR]
+
+        user32.OpenClipboard.restype = wintypes.BOOL
+        user32.OpenClipboard.argtypes = [wintypes.HWND]
+
+        user32.EmptyClipboard.restype = wintypes.BOOL
+        user32.EmptyClipboard.argtypes = []
+
+        user32.SetClipboardData.restype = wintypes.HANDLE
+        user32.SetClipboardData.argtypes = [wintypes.UINT, wintypes.HANDLE]
+
+        user32.CloseClipboard.restype = wintypes.BOOL
+        user32.CloseClipboard.argtypes = []
+
+        GMEM_MOVEABLE = 0x0002
+        CF_UNICODETEXT = 13
+
+        cf_html = user32.RegisterClipboardFormatW("HTML Format")
+        if not cf_html:
+            return False
+
+        if not user32.OpenClipboard(None):
+            return False
+
+        try:
+            user32.EmptyClipboard()
+
+            # 1. 写入 CF_HTML
+            cf_html_bytes = _format_windows_cf_html(html_str) + b"\x00"
+            h_html = kernel32.GlobalAlloc(GMEM_MOVEABLE, len(cf_html_bytes))
+            if h_html:
+                p_html = kernel32.GlobalLock(h_html)
+                if p_html:
+                    ctypes.memmove(p_html, cf_html_bytes, len(cf_html_bytes))
+                    kernel32.GlobalUnlock(h_html)
+                    user32.SetClipboardData(cf_html, h_html)
+
+            # 2. 写入 CF_UNICODETEXT (备用纯文本)
+            text_utf16 = (plain_text + "\x00").encode("utf-16le")
+            h_text = kernel32.GlobalAlloc(GMEM_MOVEABLE, len(text_utf16))
+            if h_text:
+                p_text = kernel32.GlobalLock(h_text)
+                if p_text:
+                    ctypes.memmove(p_text, text_utf16, len(text_utf16))
+                    kernel32.GlobalUnlock(h_text)
+                    user32.SetClipboardData(CF_UNICODETEXT, h_text)
+
+            return True
+        finally:
+            user32.CloseClipboard()
+    except Exception:
+        # PowerShell Set-Clipboard 作为纯文本后备方案
+        try:
+            subprocess.run(
+                ["powershell", "-NoProfile", "-Command", "Set-Clipboard", "-Value", plain_text],
+                check=True,
+                capture_output=True,
+            )
+            return True
+        except Exception:
+            return False
+
+
+def _copy_macos(html_str: str, plain_text: str) -> bool:
+    try:
+        script = f'set the clipboard to "{html_str.replace("\"", "\\\"")}"'
+        proc = subprocess.run(["osascript", "-e", script], capture_output=True)
+        return proc.returncode == 0
+    except Exception:
+        try:
+            proc = subprocess.run(["pbcopy"], input=plain_text.encode("utf-8"), capture_output=True)
+            return proc.returncode == 0
+        except Exception:
+            return False
+
+
+def _copy_linux(html_str: str, plain_text: str) -> bool:
+    try:
+        proc = subprocess.run(["xclip", "-selection", "clipboard", "-t", "text/html"], input=html_str.encode("utf-8"), capture_output=True)
+        if proc.returncode == 0:
+            return True
+    except Exception:
+        pass
+    try:
+        proc = subprocess.run(["wl-copy", "--type", "text/html"], input=html_str.encode("utf-8"), capture_output=True)
+        return proc.returncode == 0
+    except Exception:
+        return False
+
+
+def copy_html_to_clipboard(html_str: str, plain_text: Optional[str] = None) -> bool:
+    """将 HTML 富文本写入系统剪贴板。"""
+    if plain_text is None:
+        import re
+        plain_text = re.sub(r"<[^>]+>", "", html_str).strip()
+
+    system = platform.system()
+    if system == "Windows":
+        return _copy_windows(html_str, plain_text)
+    elif system == "Darwin":
+        return _copy_macos(html_str, plain_text)
+    else:
+        return _copy_linux(html_str, plain_text)

md2wx/config.py

@@ -0,0 +1,83 @@
+"""md2wx 配置与微信凭据管理模块。"""
+
+import os
+import platform
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Optional
+
+import yaml
+
+
+@dataclass
+class WeChatConfig:
+    appid: str = ""
+    secret: str = ""
+    author: str = ""
+
+
+def _get_registry_env(key: str) -> str:
+    """在 Windows 下尝试从用户注册表 HKCU\\Environment 动态读取环境变量。"""
+    if platform.system() != "Windows":
+        return ""
+    try:
+        import winreg
+        with winreg.OpenKey(winreg.HKEY_CURRENT_USER, r"Environment", 0, winreg.KEY_READ) as k:
+            val, _ = winreg.QueryValueEx(k, key)
+            return str(val).strip()
+    except Exception:
+        return ""
+
+
+def get_config_dir() -> Path:
+    p = Path.home() / ".md2wx"
+    p.mkdir(parents=True, exist_ok=True)
+    return p
+
+
+def load_config() -> dict:
+    cfg_file = get_config_dir() / "config.yaml"
+    if cfg_file.exists():
+        try:
+            data = yaml.safe_load(cfg_file.read_text(encoding="utf-8"))
+            return data if isinstance(data, dict) else {}
+        except Exception:
+            return {}
+    return {}
+
+
+def get_wechat_credentials(
+    cli_appid: Optional[str] = None,
+    cli_secret: Optional[str] = None,
+    cli_author: Optional[str] = None,
+) -> WeChatConfig:
+    """按优先级解析微信公众号凭据：CLI 参数 > 环境变量/注册表 > ~/.md2wx/config.yaml。"""
+    cfg = load_config()
+    wechat_cfg = cfg.get("wechat", {}) if isinstance(cfg.get("wechat"), dict) else {}
+
+    appid = (
+        cli_appid
+        or os.getenv("WECHAT_APPID")
+        or os.getenv("WECHAT_APP_ID")
+        or _get_registry_env("WECHAT_APPID")
+        or _get_registry_env("WECHAT_APP_ID")
+        or wechat_cfg.get("appid", "")
+    )
+
+    secret = (
+        cli_secret
+        or os.getenv("WECHAT_SECRET")
+        or os.getenv("WECHAT_APP_SECRET")
+        or _get_registry_env("WECHAT_SECRET")
+        or _get_registry_env("WECHAT_APP_SECRET")
+        or wechat_cfg.get("secret", "")
+    )
+
+    author = (
+        cli_author
+        or os.getenv("WECHAT_AUTHOR")
+        or _get_registry_env("WECHAT_AUTHOR")
+        or wechat_cfg.get("author", "")
+    )
+
+    return WeChatConfig(appid=appid.strip(), secret=secret.strip(), author=author.strip())