sophhub 0.4.19 → 0.4.21

package/skills/wechat-article-publisher/src/scripts/publish_wechat.py

@@ -1,825 +1,825 @@
-#!/usr/bin/env python3
-"""Publish Markdown/URL articles to WeChat Official Account."""
-
-from __future__ import annotations
-
-import argparse
-import json
-import os
-import re
-import subprocess
-import sys
-import textwrap
-from dataclasses import dataclass
-from pathlib import Path
-from typing import Any
-from urllib.parse import urljoin, urlparse
-
-try:
-    import markdown
-    import requests
-    import yaml
-    from bs4 import BeautifulSoup
-except Exception:
-    markdown = None
-    requests = None
-    yaml = None
-    BeautifulSoup = None
-
-TOKEN_URL = "https://api.weixin.qq.com/cgi-bin/token"
-DRAFT_ADD_URL = "https://api.weixin.qq.com/cgi-bin/draft/add"
-PUBLISH_SUBMIT_URL = "https://api.weixin.qq.com/cgi-bin/freepublish/submit"
-PUBLISH_GET_URL = "https://api.weixin.qq.com/cgi-bin/freepublish/get"
-MATERIAL_ADD_URL = "https://api.weixin.qq.com/cgi-bin/material/add_material"
-
-
-class WeChatPublishError(RuntimeError):
-    """Raised when WeChat API reports an error."""
-
-
-@dataclass
-class Article:
-    title: str
-    content: str
-    source_url: str
-    first_image_url: str
-    digest: str
-
-
-def ensure_dependencies() -> None:
-    if not all([markdown, requests, yaml, BeautifulSoup]):
-        raise RuntimeError(
-            "依赖未安装。请在项目根目录执行 uv sync，或 uv run scripts/publish_wechat.py --install"
-        )
-
-
-def install_dependencies() -> None:
-    root = Path(__file__).resolve().parent.parent
-    pyproject = root / "pyproject.toml"
-    req = Path(__file__).resolve().parent / "requirements.txt"
-    if pyproject.exists():
-        try:
-            subprocess.check_call(["uv", "sync"], cwd=str(root))
-            return
-        except FileNotFoundError:
-            pass
-    if not req.exists():
-        raise RuntimeError(f"未找到依赖文件: {req}")
-    subprocess.check_call([sys.executable, "-m", "pip", "install", "-r", str(req)])
-
-
-def slugify(text: str) -> str:
-    t = re.sub(r"[^\w\u4e00-\u9fff-]+", "-", text.strip().lower())
-    t = re.sub(r"-{2,}", "-", t).strip("-")
-    return t or "wechat-article"
-
-
-def trim_utf8_bytes(text: str, max_bytes: int) -> str:
-    raw = (text or "").encode("utf-8")
-    if len(raw) <= max_bytes:
-        return text or ""
-    return raw[:max_bytes].decode("utf-8", errors="ignore")
-
-
-def decode_escaped_unicode(text: str) -> str:
-    """Convert literal \\uXXXX sequences to real Unicode characters."""
-    if not text or "\\u" not in text:
-        return text or ""
-    if not re.search(r"\\u[0-9a-fA-F]{4}", text):
-        return text
-    try:
-        return text.encode("utf-8").decode("unicode_escape")
-    except Exception:
-        return text
-
-
-def normalize_wechat_title(text: str) -> str:
-    value = decode_escaped_unicode(text or "")
-    value = re.sub(r"\s+", " ", value.strip())
-    value = re.sub(r"^#+\s*", "", value)
-    value = value.replace("“", "\"").replace("”", "\"").replace("—", "-")
-    # 标题按字符限制：64 个汉字以内（按 64 个字符处理）。
-    value = value[:64].strip()
-    return value or "未命名文章"
-
-
-def is_url(input_value: str) -> bool:
-    return bool(re.match(r"^https?://", input_value.strip(), re.I))
-
-
-def load_config(path: Path) -> dict[str, Any]:
-    if not path.exists():
-        raise RuntimeError(f"配置文件不存在: {path}")
-    data = json.loads(path.read_text(encoding="utf-8"))
-
-    # 支持两种结构：
-    # 1) {"wechat": {...}}
-    # 2) {"platforms": {"wechat": {...}}}
-    if isinstance(data.get("wechat"), dict):
-        return data
-    if isinstance(data.get("platforms", {}).get("wechat"), dict):
-        return {"wechat": data["platforms"]["wechat"]}
-
-    raise RuntimeError("配置文件缺少 wechat 配置")
-
-
-def parse_frontmatter(md_text: str) -> tuple[dict[str, Any], str]:
-    text = decode_escaped_unicode(md_text).lstrip("\ufeff")
-    if text.startswith("---"):
-        parts = text.split("---", 2)
-        if len(parts) >= 3:
-            try:
-                frontmatter = yaml.safe_load(parts[1]) or {}
-                return frontmatter, parts[2].strip()
-            except Exception:
-                pass
-    return {}, text
-
-
-def extract_title_from_md(md_text: str) -> str:
-    text = decode_escaped_unicode(md_text).lstrip("\ufeff")
-    match = re.search(r"^#\s+(.+)$", text, re.M)
-    if match:
-        return match.group(1).strip()
-
-    for line in text.splitlines():
-        s = re.sub(r"^#+\s*", "", line.strip())
-        if s:
-            return s[:64]
-    return "未命名文章"
-
-
-def first_markdown_image(md_text: str) -> str:
-    match = re.search(r"!\[[^\]]*\]\(([^)]+)\)", md_text)
-    return match.group(1).strip() if match else ""
-
-
-def extract_from_markdown(path: Path, source_url_override: str = "") -> Article:
-    md_text = decode_escaped_unicode(path.read_text(encoding="utf-8")).lstrip("\ufeff")
-    frontmatter, body = parse_frontmatter(md_text)
-    title = (frontmatter.get("title") or extract_title_from_md(body)).strip()[:64]
-    source_url = (source_url_override or frontmatter.get("source_url") or "").strip()
-    html_content = markdown_to_html(body)
-    plain = html_to_plain_text(html_content)
-    digest = plain[:120]
-    return Article(
-        title=title or "未命名文章",
-        content=html_content,
-        source_url=source_url,
-        first_image_url=first_markdown_image(body),
-        digest=digest,
-    )
-
-
-def extract_from_url(url: str) -> Article:
-    resp = requests.get(url, timeout=30)
-    resp.raise_for_status()
-
-    soup = BeautifulSoup(resp.text, "html.parser")
-    title_node = (
-        soup.select_one("#cb_post_title_url")
-        or soup.find("h1")
-        or soup.find("title")
-    )
-    title = title_node.get_text(strip=True)[:64] if title_node else "未命名文章"
-
-    body_node = (
-        soup.select_one("#cnblogs_post_body")
-        or soup.find("article")
-        or soup.find("main")
-        or soup.find("div", class_=re.compile(r"content|article|post", re.I))
-        or soup.body
-    )
-    if not body_node:
-        raise RuntimeError("无法从网页提取正文")
-
-    for tag in body_node.select("script,style,iframe,.ad,.ads,.advertisement,.comment"):
-        tag.decompose()
-    for img in body_node.find_all("img"):
-        src = (img.get("src") or "").strip()
-        if src:
-            img["src"] = urljoin(url, src)
-    for a in body_node.find_all("a"):
-        href = (a.get("href") or "").strip()
-        if href:
-            a["href"] = urljoin(url, href)
-
-    first_image_url = ""
-    first_img = body_node.find("img")
-    if first_img and first_img.get("src"):
-        first_image_url = first_img["src"]
-
-    html_content = str(body_node)
-    plain = html_to_plain_text(html_content)
-    digest = plain[:120]
-
-    return Article(
-        title=title,
-        content=html_content,
-        source_url=url,
-        first_image_url=first_image_url,
-        digest=digest,
-    )
-
-
-def is_probably_html(text: str) -> bool:
-    sample = (text or "").strip()[:500].lower()
-    return bool(
-        re.search(r"<(article|section|div|p|h[1-6]|ul|ol|pre|table|img|blockquote)\b", sample)
-    )
-
-
-def markdown_to_html(text: str) -> str:
-    text = decode_escaped_unicode(text or "")
-    if is_probably_html(text):
-        return text
-    return markdown.markdown(
-        text,
-        extensions=["extra", "tables", "nl2br", "codehilite"],
-        output_format="html",
-    )
-
-
-def html_to_plain_text(html: str) -> str:
-    soup = BeautifulSoup(html, "html.parser")
-    txt = soup.get_text("\n", strip=True)
-    txt = re.sub(r"\n+", "\n", txt)
-    return txt
-
-
-def _next_element_sibling(node):
-    sibling = node.next_sibling
-    while sibling is not None:
-        name = getattr(sibling, "name", None)
-        if name:
-            return sibling
-        if str(sibling).strip():
-            return None
-        sibling = sibling.next_sibling
-    return None
-
-
-def _normalize_wechat_dom(soup: BeautifulSoup) -> None:
-    for list_node in soup.find_all(["ul", "ol"]):
-        for child in list(list_node.contents):
-            name = getattr(child, "name", None)
-            if name == "li":
-                continue
-            text = str(child).strip() if not name else ""
-            if text and text not in ("\n", "\r\n"):
-                li = soup.new_tag("li")
-                li.string = text
-                child.replace_with(li)
-            else:
-                try:
-                    child.extract()
-                except Exception:
-                    pass
-
-    for li in soup.find_all("li"):
-        txt = li.get_text(" ", strip=True)
-        if not txt or txt in {"•", "-", "·"}:
-            li.decompose()
-
-    for pre in soup.find_all("pre"):
-        code = pre.find("code")
-        target = code if code else pre
-        text = target.get_text() if target else ""
-        if not text:
-            continue
-        text = text.replace("\r\n", "\n").replace("\r", "\n").replace("\t", "    ")
-        lines = [line.replace("\u00a0", " ").replace("\u3000", " ").rstrip() for line in text.split("\n")]
-        cleaned = textwrap.dedent("\n".join(lines).strip("\n"))
-        target.clear()
-        target.append(cleaned)
-
-
-def _enhance_reference_links(soup: BeautifulSoup) -> None:
-    ref_keywords = ("参考资源", "相关链接", "references", "reference")
-    for heading in soup.find_all(re.compile(r"^h[1-6]$")):
-        title = heading.get_text(" ", strip=True).lower()
-        if not any(k in title for k in ref_keywords):
-            continue
-
-        current_level = int(heading.name[1])
-        node = heading.next_sibling
-        while node is not None:
-            if getattr(node, "name", None) and re.match(r"^h[1-6]$", node.name):
-                next_level = int(node.name[1])
-                if next_level <= current_level:
-                    break
-
-            anchors = node.find_all("a") if getattr(node, "find_all", None) else []
-            for a in anchors:
-                href = (a.get("href") or "").strip()
-                if not href:
-                    continue
-                parent_text = a.parent.get_text(" ", strip=True) if a.parent else ""
-                if href in parent_text:
-                    continue
-                url_anchor = soup.new_tag("a", href=href)
-                url_anchor["style"] = "color:#576b95;text-decoration:underline;word-break:break-all;"
-                url_anchor["target"] = "_blank"
-                url_anchor.string = href
-                tail = soup.new_tag("span")
-                tail["style"] = "display:block;margin-top:2px;font-size:13px;line-height:1.6;color:#7a7a7a;"
-                tail.append(url_anchor)
-                a.insert_after(tail)
-            node = node.next_sibling
-
-
-def _style_lead_paragraph(soup: BeautifulSoup, template: str) -> None:
-    for p in soup.find_all("p"):
-        if p.find_parent(["li", "blockquote", "td", "th"]):
-            continue
-        text = p.get_text(" ", strip=True)
-        if not text:
-            continue
-        if template == "viral":
-            p["style"] = (
-                "margin:1em 0;padding:0.88em 0.96em;background:#fff7e6;border:1px solid #ffe1a6;"
-                "border-radius:8px;line-height:1.9;font-size:16px;color:#2b2f38;text-indent:0;"
-                "box-shadow:0 2px 10px rgba(251,146,60,0.12);"
-            )
-        else:
-            p["style"] = (
-                "margin:1em 0;padding:0.75em 0.9em;background:#f8fafc;border:1px solid #edf2f7;"
-                "border-radius:6px;line-height:1.9;font-size:16px;color:#243044;text-indent:0;"
-            )
-        break
-
-
-def _decorate_highlights(soup: BeautifulSoup, template: str) -> None:
-    for p in soup.find_all("p"):
-        text = p.get_text(" ", strip=True)
-        if text not in ("看点：", "看点:"):
-            continue
-        if template == "viral":
-            p["style"] = (
-                "margin:0.95em 0 0.35em;display:inline-block;padding:0.24em 0.72em;"
-                "font-size:14px;line-height:1.4;color:#b42318;font-weight:700;"
-                "background:#fff1f1;border-radius:999px;border:1px solid #ffcece;"
-            )
-        else:
-            p["style"] = (
-                "margin:0.95em 0 0.35em;display:inline-block;padding:0.22em 0.65em;"
-                "font-size:14px;line-height:1.4;color:#2457c5;font-weight:700;"
-                "background:#eef4ff;border-radius:999px;border:1px solid #dce9ff;"
-            )
-
-        next_block = _next_element_sibling(p)
-        if next_block and next_block.name in ("ul", "ol"):
-            if template == "viral":
-                next_block["style"] = (
-                    "margin:0.45em 0 1em;padding:0.78em 0.98em 0.78em 1.6em;line-height:1.85;"
-                    "color:#2d3445;background:#fff7ec;border-radius:8px;"
-                    "border:1px solid #ffe3bf;box-shadow:0 2px 12px rgba(251,146,60,0.10);"
-                )
-            else:
-                next_block["style"] = (
-                    "margin:0.45em 0 1em;padding:0.75em 0.95em 0.75em 1.55em;line-height:1.82;"
-                    "color:#2f3441;background:#f8fafc;border-radius:6px;border:1px solid #e8eef8;"
-                )
-
-
-def optimize_for_wechat_html(content_html: str, template: str = "standard") -> str:
-    template = (template or "standard").strip().lower()
-    soup = BeautifulSoup(content_html, "html.parser")
-    for tag in soup.find_all(["script", "style"]):
-        tag.decompose()
-
-    _normalize_wechat_dom(soup)
-    _enhance_reference_links(soup)
-    _decorate_highlights(soup, template)
-
-    first_h1 = soup.find("h1")
-    if first_h1:
-        first_h1.decompose()
-
-    for h2 in soup.find_all("h2"):
-        if template == "viral":
-            h2["style"] = (
-                "margin:1.45em 0 0.72em;padding:0.46em 0.72em;font-size:1.2em;line-height:1.45;"
-                "color:#7a1f1f;font-weight:800;background:#fff2f2;border-left:4px solid #ef4444;border-radius:4px;"
-            )
-        else:
-            h2["style"] = (
-                "margin:1.4em 0 0.7em;padding:0.38em 0.65em;font-size:1.18em;line-height:1.45;"
-                "color:#1f2937;font-weight:700;background:#f5f8ff;border-left:4px solid #3b6dd8;border-radius:2px;"
-            )
-
-    for p in soup.find_all("p"):
-        if p.get_text(" ", strip=True) in ("看点：", "看点:"):
-            continue
-        p["style"] = (
-            "margin:0.95em 0;line-height:1.92;font-size:16px;color:#2b2f38;"
-            "text-align:justify;letter-spacing:0.01em;text-indent:0;"
-        )
-
-    list_style = (
-        "margin:0.98em 0;padding:0.82em 1em 0.82em 1.62em;line-height:1.86;color:#2d3445;"
-        "background:#fffdf8;border-radius:8px;border:1px solid #ffe8c4;"
-        "box-shadow:0 2px 10px rgba(251,146,60,0.08);list-style-position:outside;"
-    ) if template == "viral" else (
-        "margin:0.95em 0;padding:0.75em 0.95em 0.75em 1.55em;line-height:1.8;color:#2f3441;"
-        "background:#f8fafc;border-radius:6px;border:1px solid #edf2f7;list-style-position:outside;"
-    )
-
-    for ul in soup.find_all("ul"):
-        ul["style"] = list_style + "list-style-type:disc;"
-    for ol in soup.find_all("ol"):
-        ol["style"] = list_style + "list-style-type:decimal;"
-    for li in soup.find_all("li"):
-        li["style"] = "margin:0.28em 0;font-size:16px;line-height:1.82;text-indent:0;"
-
-    for a in soup.find_all("a"):
-        a["style"] = (
-            "color:#c2410c;text-decoration:underline;word-break:break-all;"
-            if template == "viral"
-            else "color:#1f57c3;text-decoration:underline;word-break:break-all;"
-        )
-        if a.get("href"):
-            a["target"] = "_blank"
-
-    for strong in soup.find_all("strong"):
-        if template == "viral":
-            strong["style"] = (
-                "font-weight:800;color:#9a3412;background:#fff3d6;padding:0 2px;border-radius:3px;"
-            )
-        else:
-            strong["style"] = "font-weight:700;color:#1f2937;"
-
-    for blockquote in soup.find_all("blockquote"):
-        if template == "viral":
-            blockquote["style"] = (
-                "margin:1.1em 0;padding:0.82em 1.02em;border-left:3px solid #f59e0b;"
-                "background:#fff8e6;color:#5f4b20;line-height:1.85;border-radius:6px;"
-            )
-        else:
-            blockquote["style"] = (
-                "margin:1.1em 0;padding:0.8em 1em;border-left:3px solid #7aa2ff;"
-                "background:#f4f7ff;color:#43506a;line-height:1.8;border-radius:4px;"
-            )
-
-    _style_lead_paragraph(soup, template)
-    body_html = "".join(str(node) for node in soup.contents)
-    wrapper = (
-        "<section style=\"margin:0 auto;max-width:690px;padding:0;"
-        "font-family:-apple-system,BlinkMacSystemFont,'Segoe UI','PingFang SC','Hiragino Sans GB','Microsoft YaHei',sans-serif;"
-        "color:#1f2937;font-size:16px;line-height:1.9;\">"
-        f"{body_html}"
-        "</section>"
-    )
-    return wrapper
-
-
-def generate_cover_image(title: str, output_path: Path) -> Path | None:
-    try:
-        from PIL import Image, ImageDraw, ImageFont
-    except Exception:
-        return None
-
-    width, height = 900, 383
-    img = Image.new("RGB", (width, height))
-    draw = ImageDraw.Draw(img)
-    top = (60, 130, 246)
-    bottom = (33, 197, 147)
-    for y in range(height):
-        ratio = y / max(1, height - 1)
-        r = int(top[0] + (bottom[0] - top[0]) * ratio)
-        g = int(top[1] + (bottom[1] - top[1]) * ratio)
-        b = int(top[2] + (bottom[2] - top[2]) * ratio)
-        draw.line([(0, y), (width, y)], fill=(r, g, b))
-
-    font_candidates = [
-        "C:\\Windows\\Fonts\\msyhbd.ttc",
-        "C:\\Windows\\Fonts\\msyh.ttc",
-        "C:\\Windows\\Fonts\\simhei.ttf",
-    ]
-    font = None
-    for c in font_candidates:
-        try:
-            font = ImageFont.truetype(c, 56)
-            break
-        except Exception:
-            continue
-    if font is None:
-        font = ImageFont.load_default()
-
-    clean = re.sub(r"\s+", " ", BeautifulSoup(title, "html.parser").get_text(" ", strip=True))
-    line = clean[:26] + ("..." if len(clean) > 26 else "")
-    draw.text((50, 130), line, fill=(255, 255, 255), font=font)
-    output_path.parent.mkdir(parents=True, exist_ok=True)
-    img.save(output_path, "JPEG", quality=92)
-    return output_path
-
-
-class WeChatClient:
-    def __init__(self, app_id: str, app_secret: str, timeout: int = 30):
-        self.app_id = app_id
-        self.app_secret = app_secret
-        self.timeout = timeout
-
-    def get_token(self) -> str:
-        resp = requests.get(
-            TOKEN_URL,
-            params={
-                "grant_type": "client_credential",
-                "appid": self.app_id,
-                "secret": self.app_secret,
-            },
-            timeout=self.timeout,
-        )
-        resp.raise_for_status()
-        data = resp.json()
-        if data.get("errcode", 0) != 0:
-            raise WeChatPublishError(f"get_token failed: {data}")
-        token = data.get("access_token", "")
-        if not token:
-            raise WeChatPublishError("get_token failed: missing access_token")
-        return token
-
-    def _post_json_utf8(self, url: str, params: dict[str, Any], payload: dict[str, Any]) -> dict[str, Any]:
-        body = json.dumps(payload, ensure_ascii=False).encode("utf-8")
-        resp = requests.post(
-            url,
-            params=params,
-            data=body,
-            headers={"Content-Type": "application/json; charset=utf-8"},
-            timeout=self.timeout,
-        )
-        resp.raise_for_status()
-        return json.loads(resp.content.decode("utf-8"))
-
-    def upload_image_from_path(self, token: str, image_path: Path) -> str:
-        with image_path.open("rb") as fh:
-            files = {"media": (image_path.name, fh, "image/jpeg")}
-            resp = requests.post(
-                MATERIAL_ADD_URL,
-                params={"access_token": token, "type": "image"},
-                files=files,
-                timeout=self.timeout,
-            )
-        resp.raise_for_status()
-        data = resp.json()
-        if data.get("errcode", 0) != 0:
-            raise WeChatPublishError(f"upload image failed: {data}")
-        media_id = data.get("media_id", "")
-        if not media_id:
-            raise WeChatPublishError("upload image failed: missing media_id")
-        return media_id
-
-    def upload_image_from_url(self, token: str, image_url: str) -> str:
-        img = requests.get(image_url, timeout=self.timeout)
-        img.raise_for_status()
-        content_type = img.headers.get("Content-Type", "image/jpeg")
-        files = {"media": ("thumb.jpg", img.content, content_type)}
-        resp = requests.post(
-            MATERIAL_ADD_URL,
-            params={"access_token": token, "type": "image"},
-            files=files,
-            timeout=self.timeout,
-        )
-        resp.raise_for_status()
-        data = resp.json()
-        if data.get("errcode", 0) != 0:
-            raise WeChatPublishError(f"upload image url failed: {data}")
-        media_id = data.get("media_id", "")
-        if not media_id:
-            raise WeChatPublishError("upload image url failed: missing media_id")
-        return media_id
-
-    def add_draft(
-        self,
-        token: str,
-        title: str,
-        author: str,
-        digest: str,
-        content_html: str,
-        source_url: str,
-        thumb_media_id: str,
-    ) -> str:
-        safe_title_base = normalize_wechat_title(title)
-        safe_author = trim_utf8_bytes(author, 16)
-        safe_digest_base = decode_escaped_unicode(digest or "")
-        safe_digest_base = re.sub(r"\s+", " ", safe_digest_base).strip()
-
-        title_limits = [64, 48, 36, 28, 20, 12]
-        digest_limits = [120, 90, 60, 40, 20]
-        title_candidates: list[str] = []
-        digest_candidates: list[str] = []
-
-        for limit in title_limits:
-            candidate = safe_title_base[:limit].strip()
-            if candidate and candidate not in title_candidates:
-                title_candidates.append(candidate)
-        for limit in digest_limits:
-            candidate = safe_digest_base[:limit].strip()
-            if candidate and candidate not in digest_candidates:
-                digest_candidates.append(candidate)
-        if not digest_candidates:
-            digest_candidates = ["AI 资讯速览"]
-
-        last_error: dict[str, Any] | None = None
-        for safe_title in title_candidates or ["未命名文章"]:
-            should_shorten_title = False
-            for safe_digest in digest_candidates:
-                payload = {
-                    "articles": [
-                        {
-                            "title": safe_title,
-                            "author": safe_author,
-                            "digest": safe_digest,
-                            "content": content_html,
-                            "content_source_url": source_url[:200],
-                            "thumb_media_id": thumb_media_id,
-                            "need_open_comment": 1,
-                            "only_fans_can_comment": 0,
-                        }
-                    ]
-                }
-                data = self._post_json_utf8(
-                    DRAFT_ADD_URL,
-                    {"access_token": token},
-                    payload,
-                )
-                if data.get("errcode", 0) == 0:
-                    media_id = data.get("media_id", "")
-                    if not media_id:
-                        raise WeChatPublishError("draft add failed: missing media_id")
-                    return media_id
-
-                last_error = data
-                errcode = data.get("errcode")
-                if errcode == 45003:
-                    should_shorten_title = True
-                    break
-                if errcode == 45004:
-                    continue
-                raise WeChatPublishError(f"draft add failed: {data}")
-
-            if should_shorten_title:
-                continue
-
-        raise WeChatPublishError(f"draft add failed: {last_error}")
-
-    def submit_publish(self, token: str, media_id: str) -> str:
-        data = self._post_json_utf8(
-            PUBLISH_SUBMIT_URL,
-            {"access_token": token},
-            {"media_id": media_id},
-        )
-        if data.get("errcode", 0) != 0:
-            raise WeChatPublishError(f"publish submit failed: {data}")
-        publish_id = data.get("publish_id", "")
-        if not publish_id:
-            raise WeChatPublishError("publish submit failed: missing publish_id")
-        return publish_id
-
-    def get_publish_status(self, token: str, publish_id: str) -> dict[str, Any]:
-        data = self._post_json_utf8(
-            PUBLISH_GET_URL,
-            {"access_token": token},
-            {"publish_id": publish_id},
-        )
-        if data.get("errcode", 0) != 0:
-            raise WeChatPublishError(f"publish get failed: {data}")
-        return data
-
-
-def parse_args() -> argparse.Namespace:
-    default_config = Path(__file__).resolve().parent.parent / "config.json"
-    parser = argparse.ArgumentParser(description="Publish article to WeChat Official Account")
-    parser.add_argument("input", nargs="?", help="Markdown file path or article URL")
-    parser.add_argument("--config", default=str(default_config), help="Path to config.json")
-    parser.add_argument("--template", choices=["standard", "viral"], default="", help="Override template")
-    parser.add_argument("--author", default="", help="Override author")
-    parser.add_argument("--source-url", default="", help="Override source url")
-    parser.add_argument("--cover-image", default="", help="Local cover image path")
-    parser.add_argument("--timeout", type=int, default=30, help="HTTP timeout seconds")
-    parser.add_argument("--dry-run", action="store_true", help="Render only, no WeChat API calls")
-    parser.add_argument("--publish", action="store_true", help="Submit draft for publish")
-    parser.add_argument("--status", action="store_true", help="Query publish status once")
-    parser.add_argument("--install", action="store_true", help="Install Python dependencies")
-    return parser.parse_args()
-
-
-def main() -> None:
-    args = parse_args()
-    if args.install:
-        install_dependencies()
-        print(json.dumps({"success": True, "installed": True}, ensure_ascii=False))
-        return
-
-    if not args.input:
-        raise RuntimeError("缺少输入参数：请传入 Markdown 文件路径或 URL")
-    ensure_dependencies()
-
-    cfg = load_config(Path(args.config))
-    wechat_cfg = cfg["wechat"]
-    template = (args.template or "viral").strip().lower()
-    if template not in {"standard", "viral"}:
-        template = "standard"
-    author = (args.author or wechat_cfg.get("author") or "").strip()
-    input_value = args.input.strip()
-
-    if is_url(input_value):
-        article = extract_from_url(input_value)
-        if args.source_url:
-            article.source_url = args.source_url.strip()
-    else:
-        input_path = Path(input_value).resolve()
-        if not input_path.exists():
-            raise RuntimeError(f"文件不存在: {input_path}")
-        article = extract_from_markdown(input_path, source_url_override=args.source_url.strip())
-
-    article.content = optimize_for_wechat_html(article.content, template=template)
-
-    preview_path = Path.cwd() / f"{slugify(article.title)}-wechat-preview.html"
-    preview_path.write_text(article.content, encoding="utf-8")
-
-    if args.dry_run:
-        print(
-            json.dumps(
-                {
-                    "success": True,
-                    "mode": "dry-run",
-                    "title": article.title,
-                    "digest": article.digest,
-                    "template": template,
-                    "preview_html": str(preview_path),
-                },
-                ensure_ascii=False,
-            )
-        )
-        return
-
-    app_id = (wechat_cfg.get("app_id") or "").strip()
-    app_secret = (wechat_cfg.get("app_secret") or "").strip()
-    if not app_id or not app_secret:
-        raise RuntimeError("配置缺少 wechat.app_id 或 wechat.app_secret")
-
-    client = WeChatClient(app_id=app_id, app_secret=app_secret, timeout=args.timeout)
-    token = client.get_token()
-
-    thumb_media_id = ""
-    auto_generate_cover = True
-    cover_image_path = Path(args.cover_image).resolve() if args.cover_image else None
-    if not thumb_media_id:
-        if cover_image_path and cover_image_path.exists():
-            thumb_media_id = client.upload_image_from_path(token, cover_image_path)
-        else:
-            generated = None
-            if auto_generate_cover:
-                generated = generate_cover_image(
-                    article.title,
-                    Path(__file__).resolve().parent.parent / "assets" / "generated_cover.jpg",
-                )
-            if generated and generated.exists():
-                thumb_media_id = client.upload_image_from_path(token, generated)
-            elif article.first_image_url and article.first_image_url.startswith(("http://", "https://")):
-                thumb_media_id = client.upload_image_from_url(token, article.first_image_url)
-
-    if not thumb_media_id:
-        raise RuntimeError(
-            "无法获取封面素材，请提供 --cover-image，或确保正文含可下载图片，或安装 Pillow 以自动生成封面"
-        )
-
-    draft_media_id = client.add_draft(
-        token=token,
-        title=article.title,
-        author=author,
-        digest=article.digest,
-        content_html=article.content,
-        source_url=article.source_url,
-        thumb_media_id=thumb_media_id,
-    )
-
-    result: dict[str, Any] = {
-        "success": True,
-        "title": article.title,
-        "template": template,
-        "draft_media_id": draft_media_id,
-        "preview_html": str(preview_path),
-    }
-
-    if args.publish:
-        publish_id = client.submit_publish(token, draft_media_id)
-        result["publish_id"] = publish_id
-        if args.status:
-            result["status"] = client.get_publish_status(token, publish_id)
-
-    print(json.dumps(result, ensure_ascii=False))
-
-
-if __name__ == "__main__":
-    try:
-        main()
-    except Exception as exc:
-        print(
-            json.dumps(
-                {"success": False, "error": str(exc), "type": type(exc).__name__},
-                ensure_ascii=False,
-            )
-        )
-        sys.exit(1)
+#!/usr/bin/env python3
+"""Publish Markdown/URL articles to WeChat Official Account."""
+
+from __future__ import annotations
+
+import argparse
+import json
+import os
+import re
+import subprocess
+import sys
+import textwrap
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any
+from urllib.parse import urljoin, urlparse
+
+try:
+    import markdown
+    import requests
+    import yaml
+    from bs4 import BeautifulSoup
+except Exception:
+    markdown = None
+    requests = None
+    yaml = None
+    BeautifulSoup = None
+
+TOKEN_URL = "https://api.weixin.qq.com/cgi-bin/token"
+DRAFT_ADD_URL = "https://api.weixin.qq.com/cgi-bin/draft/add"
+PUBLISH_SUBMIT_URL = "https://api.weixin.qq.com/cgi-bin/freepublish/submit"
+PUBLISH_GET_URL = "https://api.weixin.qq.com/cgi-bin/freepublish/get"
+MATERIAL_ADD_URL = "https://api.weixin.qq.com/cgi-bin/material/add_material"
+
+
+class WeChatPublishError(RuntimeError):
+    """Raised when WeChat API reports an error."""
+
+
+@dataclass
+class Article:
+    title: str
+    content: str
+    source_url: str
+    first_image_url: str
+    digest: str
+
+
+def ensure_dependencies() -> None:
+    if not all([markdown, requests, yaml, BeautifulSoup]):
+        raise RuntimeError(
+            "依赖未安装。请在项目根目录执行 uv sync，或 uv run scripts/publish_wechat.py --install"
+        )
+
+
+def install_dependencies() -> None:
+    root = Path(__file__).resolve().parent.parent
+    pyproject = root / "pyproject.toml"
+    req = Path(__file__).resolve().parent / "requirements.txt"
+    if pyproject.exists():
+        try:
+            subprocess.check_call(["uv", "sync"], cwd=str(root))
+            return
+        except FileNotFoundError:
+            pass
+    if not req.exists():
+        raise RuntimeError(f"未找到依赖文件: {req}")
+    subprocess.check_call([sys.executable, "-m", "pip", "install", "-r", str(req)])
+
+
+def slugify(text: str) -> str:
+    t = re.sub(r"[^\w\u4e00-\u9fff-]+", "-", text.strip().lower())
+    t = re.sub(r"-{2,}", "-", t).strip("-")
+    return t or "wechat-article"
+
+
+def trim_utf8_bytes(text: str, max_bytes: int) -> str:
+    raw = (text or "").encode("utf-8")
+    if len(raw) <= max_bytes:
+        return text or ""
+    return raw[:max_bytes].decode("utf-8", errors="ignore")
+
+
+def decode_escaped_unicode(text: str) -> str:
+    """Convert literal \\uXXXX sequences to real Unicode characters."""
+    if not text or "\\u" not in text:
+        return text or ""
+    if not re.search(r"\\u[0-9a-fA-F]{4}", text):
+        return text
+    try:
+        return text.encode("utf-8").decode("unicode_escape")
+    except Exception:
+        return text
+
+
+def normalize_wechat_title(text: str) -> str:
+    value = decode_escaped_unicode(text or "")
+    value = re.sub(r"\s+", " ", value.strip())
+    value = re.sub(r"^#+\s*", "", value)
+    value = value.replace("“", "\"").replace("”", "\"").replace("—", "-")
+    # 标题按字符限制：64 个汉字以内（按 64 个字符处理）。
+    value = value[:64].strip()
+    return value or "未命名文章"
+
+
+def is_url(input_value: str) -> bool:
+    return bool(re.match(r"^https?://", input_value.strip(), re.I))
+
+
+def load_config(path: Path) -> dict[str, Any]:
+    if not path.exists():
+        raise RuntimeError(f"配置文件不存在: {path}")
+    data = json.loads(path.read_text(encoding="utf-8"))
+
+    # 支持两种结构：
+    # 1) {"wechat": {...}}
+    # 2) {"platforms": {"wechat": {...}}}
+    if isinstance(data.get("wechat"), dict):
+        return data
+    if isinstance(data.get("platforms", {}).get("wechat"), dict):
+        return {"wechat": data["platforms"]["wechat"]}
+
+    raise RuntimeError("配置文件缺少 wechat 配置")
+
+
+def parse_frontmatter(md_text: str) -> tuple[dict[str, Any], str]:
+    text = decode_escaped_unicode(md_text).lstrip("\ufeff")
+    if text.startswith("---"):
+        parts = text.split("---", 2)
+        if len(parts) >= 3:
+            try:
+                frontmatter = yaml.safe_load(parts[1]) or {}
+                return frontmatter, parts[2].strip()
+            except Exception:
+                pass
+    return {}, text
+
+
+def extract_title_from_md(md_text: str) -> str:
+    text = decode_escaped_unicode(md_text).lstrip("\ufeff")
+    match = re.search(r"^#\s+(.+)$", text, re.M)
+    if match:
+        return match.group(1).strip()
+
+    for line in text.splitlines():
+        s = re.sub(r"^#+\s*", "", line.strip())
+        if s:
+            return s[:64]
+    return "未命名文章"
+
+
+def first_markdown_image(md_text: str) -> str:
+    match = re.search(r"!\[[^\]]*\]\(([^)]+)\)", md_text)
+    return match.group(1).strip() if match else ""
+
+
+def extract_from_markdown(path: Path, source_url_override: str = "") -> Article:
+    md_text = decode_escaped_unicode(path.read_text(encoding="utf-8")).lstrip("\ufeff")
+    frontmatter, body = parse_frontmatter(md_text)
+    title = (frontmatter.get("title") or extract_title_from_md(body)).strip()[:64]
+    source_url = (source_url_override or frontmatter.get("source_url") or "").strip()
+    html_content = markdown_to_html(body)
+    plain = html_to_plain_text(html_content)
+    digest = plain[:120]
+    return Article(
+        title=title or "未命名文章",
+        content=html_content,
+        source_url=source_url,
+        first_image_url=first_markdown_image(body),
+        digest=digest,
+    )
+
+
+def extract_from_url(url: str) -> Article:
+    resp = requests.get(url, timeout=30)
+    resp.raise_for_status()
+
+    soup = BeautifulSoup(resp.text, "html.parser")
+    title_node = (
+        soup.select_one("#cb_post_title_url")
+        or soup.find("h1")
+        or soup.find("title")
+    )
+    title = title_node.get_text(strip=True)[:64] if title_node else "未命名文章"
+
+    body_node = (
+        soup.select_one("#cnblogs_post_body")
+        or soup.find("article")
+        or soup.find("main")
+        or soup.find("div", class_=re.compile(r"content|article|post", re.I))
+        or soup.body
+    )
+    if not body_node:
+        raise RuntimeError("无法从网页提取正文")
+
+    for tag in body_node.select("script,style,iframe,.ad,.ads,.advertisement,.comment"):
+        tag.decompose()
+    for img in body_node.find_all("img"):
+        src = (img.get("src") or "").strip()
+        if src:
+            img["src"] = urljoin(url, src)
+    for a in body_node.find_all("a"):
+        href = (a.get("href") or "").strip()
+        if href:
+            a["href"] = urljoin(url, href)
+
+    first_image_url = ""
+    first_img = body_node.find("img")
+    if first_img and first_img.get("src"):
+        first_image_url = first_img["src"]
+
+    html_content = str(body_node)
+    plain = html_to_plain_text(html_content)
+    digest = plain[:120]
+
+    return Article(
+        title=title,
+        content=html_content,
+        source_url=url,
+        first_image_url=first_image_url,
+        digest=digest,
+    )
+
+
+def is_probably_html(text: str) -> bool:
+    sample = (text or "").strip()[:500].lower()
+    return bool(
+        re.search(r"<(article|section|div|p|h[1-6]|ul|ol|pre|table|img|blockquote)\b", sample)
+    )
+
+
+def markdown_to_html(text: str) -> str:
+    text = decode_escaped_unicode(text or "")
+    if is_probably_html(text):
+        return text
+    return markdown.markdown(
+        text,
+        extensions=["extra", "tables", "nl2br", "codehilite"],
+        output_format="html",
+    )
+
+
+def html_to_plain_text(html: str) -> str:
+    soup = BeautifulSoup(html, "html.parser")
+    txt = soup.get_text("\n", strip=True)
+    txt = re.sub(r"\n+", "\n", txt)
+    return txt
+
+
+def _next_element_sibling(node):
+    sibling = node.next_sibling
+    while sibling is not None:
+        name = getattr(sibling, "name", None)
+        if name:
+            return sibling
+        if str(sibling).strip():
+            return None
+        sibling = sibling.next_sibling
+    return None
+
+
+def _normalize_wechat_dom(soup: BeautifulSoup) -> None:
+    for list_node in soup.find_all(["ul", "ol"]):
+        for child in list(list_node.contents):
+            name = getattr(child, "name", None)
+            if name == "li":
+                continue
+            text = str(child).strip() if not name else ""
+            if text and text not in ("\n", "\r\n"):
+                li = soup.new_tag("li")
+                li.string = text
+                child.replace_with(li)
+            else:
+                try:
+                    child.extract()
+                except Exception:
+                    pass
+
+    for li in soup.find_all("li"):
+        txt = li.get_text(" ", strip=True)
+        if not txt or txt in {"•", "-", "·"}:
+            li.decompose()
+
+    for pre in soup.find_all("pre"):
+        code = pre.find("code")
+        target = code if code else pre
+        text = target.get_text() if target else ""
+        if not text:
+            continue
+        text = text.replace("\r\n", "\n").replace("\r", "\n").replace("\t", "    ")
+        lines = [line.replace("\u00a0", " ").replace("\u3000", " ").rstrip() for line in text.split("\n")]
+        cleaned = textwrap.dedent("\n".join(lines).strip("\n"))
+        target.clear()
+        target.append(cleaned)
+
+
+def _enhance_reference_links(soup: BeautifulSoup) -> None:
+    ref_keywords = ("参考资源", "相关链接", "references", "reference")
+    for heading in soup.find_all(re.compile(r"^h[1-6]$")):
+        title = heading.get_text(" ", strip=True).lower()
+        if not any(k in title for k in ref_keywords):
+            continue
+
+        current_level = int(heading.name[1])
+        node = heading.next_sibling
+        while node is not None:
+            if getattr(node, "name", None) and re.match(r"^h[1-6]$", node.name):
+                next_level = int(node.name[1])
+                if next_level <= current_level:
+                    break
+
+            anchors = node.find_all("a") if getattr(node, "find_all", None) else []
+            for a in anchors:
+                href = (a.get("href") or "").strip()
+                if not href:
+                    continue
+                parent_text = a.parent.get_text(" ", strip=True) if a.parent else ""
+                if href in parent_text:
+                    continue
+                url_anchor = soup.new_tag("a", href=href)
+                url_anchor["style"] = "color:#576b95;text-decoration:underline;word-break:break-all;"
+                url_anchor["target"] = "_blank"
+                url_anchor.string = href
+                tail = soup.new_tag("span")
+                tail["style"] = "display:block;margin-top:2px;font-size:13px;line-height:1.6;color:#7a7a7a;"
+                tail.append(url_anchor)
+                a.insert_after(tail)
+            node = node.next_sibling
+
+
+def _style_lead_paragraph(soup: BeautifulSoup, template: str) -> None:
+    for p in soup.find_all("p"):
+        if p.find_parent(["li", "blockquote", "td", "th"]):
+            continue
+        text = p.get_text(" ", strip=True)
+        if not text:
+            continue
+        if template == "viral":
+            p["style"] = (
+                "margin:1em 0;padding:0.88em 0.96em;background:#fff7e6;border:1px solid #ffe1a6;"
+                "border-radius:8px;line-height:1.9;font-size:16px;color:#2b2f38;text-indent:0;"
+                "box-shadow:0 2px 10px rgba(251,146,60,0.12);"
+            )
+        else:
+            p["style"] = (
+                "margin:1em 0;padding:0.75em 0.9em;background:#f8fafc;border:1px solid #edf2f7;"
+                "border-radius:6px;line-height:1.9;font-size:16px;color:#243044;text-indent:0;"
+            )
+        break
+
+
+def _decorate_highlights(soup: BeautifulSoup, template: str) -> None:
+    for p in soup.find_all("p"):
+        text = p.get_text(" ", strip=True)
+        if text not in ("看点：", "看点:"):
+            continue
+        if template == "viral":
+            p["style"] = (
+                "margin:0.95em 0 0.35em;display:inline-block;padding:0.24em 0.72em;"
+                "font-size:14px;line-height:1.4;color:#b42318;font-weight:700;"
+                "background:#fff1f1;border-radius:999px;border:1px solid #ffcece;"
+            )
+        else:
+            p["style"] = (
+                "margin:0.95em 0 0.35em;display:inline-block;padding:0.22em 0.65em;"
+                "font-size:14px;line-height:1.4;color:#2457c5;font-weight:700;"
+                "background:#eef4ff;border-radius:999px;border:1px solid #dce9ff;"
+            )
+
+        next_block = _next_element_sibling(p)
+        if next_block and next_block.name in ("ul", "ol"):
+            if template == "viral":
+                next_block["style"] = (
+                    "margin:0.45em 0 1em;padding:0.78em 0.98em 0.78em 1.6em;line-height:1.85;"
+                    "color:#2d3445;background:#fff7ec;border-radius:8px;"
+                    "border:1px solid #ffe3bf;box-shadow:0 2px 12px rgba(251,146,60,0.10);"
+                )
+            else:
+                next_block["style"] = (
+                    "margin:0.45em 0 1em;padding:0.75em 0.95em 0.75em 1.55em;line-height:1.82;"
+                    "color:#2f3441;background:#f8fafc;border-radius:6px;border:1px solid #e8eef8;"
+                )
+
+
+def optimize_for_wechat_html(content_html: str, template: str = "standard") -> str:
+    template = (template or "standard").strip().lower()
+    soup = BeautifulSoup(content_html, "html.parser")
+    for tag in soup.find_all(["script", "style"]):
+        tag.decompose()
+
+    _normalize_wechat_dom(soup)
+    _enhance_reference_links(soup)
+    _decorate_highlights(soup, template)
+
+    first_h1 = soup.find("h1")
+    if first_h1:
+        first_h1.decompose()
+
+    for h2 in soup.find_all("h2"):
+        if template == "viral":
+            h2["style"] = (
+                "margin:1.45em 0 0.72em;padding:0.46em 0.72em;font-size:1.2em;line-height:1.45;"
+                "color:#7a1f1f;font-weight:800;background:#fff2f2;border-left:4px solid #ef4444;border-radius:4px;"
+            )
+        else:
+            h2["style"] = (
+                "margin:1.4em 0 0.7em;padding:0.38em 0.65em;font-size:1.18em;line-height:1.45;"
+                "color:#1f2937;font-weight:700;background:#f5f8ff;border-left:4px solid #3b6dd8;border-radius:2px;"
+            )
+
+    for p in soup.find_all("p"):
+        if p.get_text(" ", strip=True) in ("看点：", "看点:"):
+            continue
+        p["style"] = (
+            "margin:0.95em 0;line-height:1.92;font-size:16px;color:#2b2f38;"
+            "text-align:justify;letter-spacing:0.01em;text-indent:0;"
+        )
+
+    list_style = (
+        "margin:0.98em 0;padding:0.82em 1em 0.82em 1.62em;line-height:1.86;color:#2d3445;"
+        "background:#fffdf8;border-radius:8px;border:1px solid #ffe8c4;"
+        "box-shadow:0 2px 10px rgba(251,146,60,0.08);list-style-position:outside;"
+    ) if template == "viral" else (
+        "margin:0.95em 0;padding:0.75em 0.95em 0.75em 1.55em;line-height:1.8;color:#2f3441;"
+        "background:#f8fafc;border-radius:6px;border:1px solid #edf2f7;list-style-position:outside;"
+    )
+
+    for ul in soup.find_all("ul"):
+        ul["style"] = list_style + "list-style-type:disc;"
+    for ol in soup.find_all("ol"):
+        ol["style"] = list_style + "list-style-type:decimal;"
+    for li in soup.find_all("li"):
+        li["style"] = "margin:0.28em 0;font-size:16px;line-height:1.82;text-indent:0;"
+
+    for a in soup.find_all("a"):
+        a["style"] = (
+            "color:#c2410c;text-decoration:underline;word-break:break-all;"
+            if template == "viral"
+            else "color:#1f57c3;text-decoration:underline;word-break:break-all;"
+        )
+        if a.get("href"):
+            a["target"] = "_blank"
+
+    for strong in soup.find_all("strong"):
+        if template == "viral":
+            strong["style"] = (
+                "font-weight:800;color:#9a3412;background:#fff3d6;padding:0 2px;border-radius:3px;"
+            )
+        else:
+            strong["style"] = "font-weight:700;color:#1f2937;"
+
+    for blockquote in soup.find_all("blockquote"):
+        if template == "viral":
+            blockquote["style"] = (
+                "margin:1.1em 0;padding:0.82em 1.02em;border-left:3px solid #f59e0b;"
+                "background:#fff8e6;color:#5f4b20;line-height:1.85;border-radius:6px;"
+            )
+        else:
+            blockquote["style"] = (
+                "margin:1.1em 0;padding:0.8em 1em;border-left:3px solid #7aa2ff;"
+                "background:#f4f7ff;color:#43506a;line-height:1.8;border-radius:4px;"
+            )
+
+    _style_lead_paragraph(soup, template)
+    body_html = "".join(str(node) for node in soup.contents)
+    wrapper = (
+        "<section style=\"margin:0 auto;max-width:690px;padding:0;"
+        "font-family:-apple-system,BlinkMacSystemFont,'Segoe UI','PingFang SC','Hiragino Sans GB','Microsoft YaHei',sans-serif;"
+        "color:#1f2937;font-size:16px;line-height:1.9;\">"
+        f"{body_html}"
+        "</section>"
+    )
+    return wrapper
+
+
+def generate_cover_image(title: str, output_path: Path) -> Path | None:
+    try:
+        from PIL import Image, ImageDraw, ImageFont
+    except Exception:
+        return None
+
+    width, height = 900, 383
+    img = Image.new("RGB", (width, height))
+    draw = ImageDraw.Draw(img)
+    top = (60, 130, 246)
+    bottom = (33, 197, 147)
+    for y in range(height):
+        ratio = y / max(1, height - 1)
+        r = int(top[0] + (bottom[0] - top[0]) * ratio)
+        g = int(top[1] + (bottom[1] - top[1]) * ratio)
+        b = int(top[2] + (bottom[2] - top[2]) * ratio)
+        draw.line([(0, y), (width, y)], fill=(r, g, b))
+
+    font_candidates = [
+        "C:\\Windows\\Fonts\\msyhbd.ttc",
+        "C:\\Windows\\Fonts\\msyh.ttc",
+        "C:\\Windows\\Fonts\\simhei.ttf",
+    ]
+    font = None
+    for c in font_candidates:
+        try:
+            font = ImageFont.truetype(c, 56)
+            break
+        except Exception:
+            continue
+    if font is None:
+        font = ImageFont.load_default()
+
+    clean = re.sub(r"\s+", " ", BeautifulSoup(title, "html.parser").get_text(" ", strip=True))
+    line = clean[:26] + ("..." if len(clean) > 26 else "")
+    draw.text((50, 130), line, fill=(255, 255, 255), font=font)
+    output_path.parent.mkdir(parents=True, exist_ok=True)
+    img.save(output_path, "JPEG", quality=92)
+    return output_path
+
+
+class WeChatClient:
+    def __init__(self, app_id: str, app_secret: str, timeout: int = 30):
+        self.app_id = app_id
+        self.app_secret = app_secret
+        self.timeout = timeout
+
+    def get_token(self) -> str:
+        resp = requests.get(
+            TOKEN_URL,
+            params={
+                "grant_type": "client_credential",
+                "appid": self.app_id,
+                "secret": self.app_secret,
+            },
+            timeout=self.timeout,
+        )
+        resp.raise_for_status()
+        data = resp.json()
+        if data.get("errcode", 0) != 0:
+            raise WeChatPublishError(f"get_token failed: {data}")
+        token = data.get("access_token", "")
+        if not token:
+            raise WeChatPublishError("get_token failed: missing access_token")
+        return token
+
+    def _post_json_utf8(self, url: str, params: dict[str, Any], payload: dict[str, Any]) -> dict[str, Any]:
+        body = json.dumps(payload, ensure_ascii=False).encode("utf-8")
+        resp = requests.post(
+            url,
+            params=params,
+            data=body,
+            headers={"Content-Type": "application/json; charset=utf-8"},
+            timeout=self.timeout,
+        )
+        resp.raise_for_status()
+        return json.loads(resp.content.decode("utf-8"))
+
+    def upload_image_from_path(self, token: str, image_path: Path) -> str:
+        with image_path.open("rb") as fh:
+            files = {"media": (image_path.name, fh, "image/jpeg")}
+            resp = requests.post(
+                MATERIAL_ADD_URL,
+                params={"access_token": token, "type": "image"},
+                files=files,
+                timeout=self.timeout,
+            )
+        resp.raise_for_status()
+        data = resp.json()
+        if data.get("errcode", 0) != 0:
+            raise WeChatPublishError(f"upload image failed: {data}")
+        media_id = data.get("media_id", "")
+        if not media_id:
+            raise WeChatPublishError("upload image failed: missing media_id")
+        return media_id
+
+    def upload_image_from_url(self, token: str, image_url: str) -> str:
+        img = requests.get(image_url, timeout=self.timeout)
+        img.raise_for_status()
+        content_type = img.headers.get("Content-Type", "image/jpeg")
+        files = {"media": ("thumb.jpg", img.content, content_type)}
+        resp = requests.post(
+            MATERIAL_ADD_URL,
+            params={"access_token": token, "type": "image"},
+            files=files,
+            timeout=self.timeout,
+        )
+        resp.raise_for_status()
+        data = resp.json()
+        if data.get("errcode", 0) != 0:
+            raise WeChatPublishError(f"upload image url failed: {data}")
+        media_id = data.get("media_id", "")
+        if not media_id:
+            raise WeChatPublishError("upload image url failed: missing media_id")
+        return media_id
+
+    def add_draft(
+        self,
+        token: str,
+        title: str,
+        author: str,
+        digest: str,
+        content_html: str,
+        source_url: str,
+        thumb_media_id: str,
+    ) -> str:
+        safe_title_base = normalize_wechat_title(title)
+        safe_author = trim_utf8_bytes(author, 16)
+        safe_digest_base = decode_escaped_unicode(digest or "")
+        safe_digest_base = re.sub(r"\s+", " ", safe_digest_base).strip()
+
+        title_limits = [64, 48, 36, 28, 20, 12]
+        digest_limits = [120, 90, 60, 40, 20]
+        title_candidates: list[str] = []
+        digest_candidates: list[str] = []
+
+        for limit in title_limits:
+            candidate = safe_title_base[:limit].strip()
+            if candidate and candidate not in title_candidates:
+                title_candidates.append(candidate)
+        for limit in digest_limits:
+            candidate = safe_digest_base[:limit].strip()
+            if candidate and candidate not in digest_candidates:
+                digest_candidates.append(candidate)
+        if not digest_candidates:
+            digest_candidates = ["AI 资讯速览"]
+
+        last_error: dict[str, Any] | None = None
+        for safe_title in title_candidates or ["未命名文章"]:
+            should_shorten_title = False
+            for safe_digest in digest_candidates:
+                payload = {
+                    "articles": [
+                        {
+                            "title": safe_title,
+                            "author": safe_author,
+                            "digest": safe_digest,
+                            "content": content_html,
+                            "content_source_url": source_url[:200],
+                            "thumb_media_id": thumb_media_id,
+                            "need_open_comment": 1,
+                            "only_fans_can_comment": 0,
+                        }
+                    ]
+                }
+                data = self._post_json_utf8(
+                    DRAFT_ADD_URL,
+                    {"access_token": token},
+                    payload,
+                )
+                if data.get("errcode", 0) == 0:
+                    media_id = data.get("media_id", "")
+                    if not media_id:
+                        raise WeChatPublishError("draft add failed: missing media_id")
+                    return media_id
+
+                last_error = data
+                errcode = data.get("errcode")
+                if errcode == 45003:
+                    should_shorten_title = True
+                    break
+                if errcode == 45004:
+                    continue
+                raise WeChatPublishError(f"draft add failed: {data}")
+
+            if should_shorten_title:
+                continue
+
+        raise WeChatPublishError(f"draft add failed: {last_error}")
+
+    def submit_publish(self, token: str, media_id: str) -> str:
+        data = self._post_json_utf8(
+            PUBLISH_SUBMIT_URL,
+            {"access_token": token},
+            {"media_id": media_id},
+        )
+        if data.get("errcode", 0) != 0:
+            raise WeChatPublishError(f"publish submit failed: {data}")
+        publish_id = data.get("publish_id", "")
+        if not publish_id:
+            raise WeChatPublishError("publish submit failed: missing publish_id")
+        return publish_id
+
+    def get_publish_status(self, token: str, publish_id: str) -> dict[str, Any]:
+        data = self._post_json_utf8(
+            PUBLISH_GET_URL,
+            {"access_token": token},
+            {"publish_id": publish_id},
+        )
+        if data.get("errcode", 0) != 0:
+            raise WeChatPublishError(f"publish get failed: {data}")
+        return data
+
+
+def parse_args() -> argparse.Namespace:
+    default_config = Path(__file__).resolve().parent.parent / "config.json"
+    parser = argparse.ArgumentParser(description="Publish article to WeChat Official Account")
+    parser.add_argument("input", nargs="?", help="Markdown file path or article URL")
+    parser.add_argument("--config", default=str(default_config), help="Path to config.json")
+    parser.add_argument("--template", choices=["standard", "viral"], default="", help="Override template")
+    parser.add_argument("--author", default="", help="Override author")
+    parser.add_argument("--source-url", default="", help="Override source url")
+    parser.add_argument("--cover-image", default="", help="Local cover image path")
+    parser.add_argument("--timeout", type=int, default=30, help="HTTP timeout seconds")
+    parser.add_argument("--dry-run", action="store_true", help="Render only, no WeChat API calls")
+    parser.add_argument("--publish", action="store_true", help="Submit draft for publish")
+    parser.add_argument("--status", action="store_true", help="Query publish status once")
+    parser.add_argument("--install", action="store_true", help="Install Python dependencies")
+    return parser.parse_args()
+
+
+def main() -> None:
+    args = parse_args()
+    if args.install:
+        install_dependencies()
+        print(json.dumps({"success": True, "installed": True}, ensure_ascii=False))
+        return
+
+    if not args.input:
+        raise RuntimeError("缺少输入参数：请传入 Markdown 文件路径或 URL")
+    ensure_dependencies()
+
+    cfg = load_config(Path(args.config))
+    wechat_cfg = cfg["wechat"]
+    template = (args.template or "viral").strip().lower()
+    if template not in {"standard", "viral"}:
+        template = "standard"
+    author = (args.author or wechat_cfg.get("author") or "").strip()
+    input_value = args.input.strip()
+
+    if is_url(input_value):
+        article = extract_from_url(input_value)
+        if args.source_url:
+            article.source_url = args.source_url.strip()
+    else:
+        input_path = Path(input_value).resolve()
+        if not input_path.exists():
+            raise RuntimeError(f"文件不存在: {input_path}")
+        article = extract_from_markdown(input_path, source_url_override=args.source_url.strip())
+
+    article.content = optimize_for_wechat_html(article.content, template=template)
+
+    preview_path = Path.cwd() / f"{slugify(article.title)}-wechat-preview.html"
+    preview_path.write_text(article.content, encoding="utf-8")
+
+    if args.dry_run:
+        print(
+            json.dumps(
+                {
+                    "success": True,
+                    "mode": "dry-run",
+                    "title": article.title,
+                    "digest": article.digest,
+                    "template": template,
+                    "preview_html": str(preview_path),
+                },
+                ensure_ascii=False,
+            )
+        )
+        return
+
+    app_id = (wechat_cfg.get("app_id") or "").strip()
+    app_secret = (wechat_cfg.get("app_secret") or "").strip()
+    if not app_id or not app_secret:
+        raise RuntimeError("配置缺少 wechat.app_id 或 wechat.app_secret")
+
+    client = WeChatClient(app_id=app_id, app_secret=app_secret, timeout=args.timeout)
+    token = client.get_token()
+
+    thumb_media_id = ""
+    auto_generate_cover = True
+    cover_image_path = Path(args.cover_image).resolve() if args.cover_image else None
+    if not thumb_media_id:
+        if cover_image_path and cover_image_path.exists():
+            thumb_media_id = client.upload_image_from_path(token, cover_image_path)
+        else:
+            generated = None
+            if auto_generate_cover:
+                generated = generate_cover_image(
+                    article.title,
+                    Path(__file__).resolve().parent.parent / "assets" / "generated_cover.jpg",
+                )
+            if generated and generated.exists():
+                thumb_media_id = client.upload_image_from_path(token, generated)
+            elif article.first_image_url and article.first_image_url.startswith(("http://", "https://")):
+                thumb_media_id = client.upload_image_from_url(token, article.first_image_url)
+
+    if not thumb_media_id:
+        raise RuntimeError(
+            "无法获取封面素材，请提供 --cover-image，或确保正文含可下载图片，或安装 Pillow 以自动生成封面"
+        )
+
+    draft_media_id = client.add_draft(
+        token=token,
+        title=article.title,
+        author=author,
+        digest=article.digest,
+        content_html=article.content,
+        source_url=article.source_url,
+        thumb_media_id=thumb_media_id,
+    )
+
+    result: dict[str, Any] = {
+        "success": True,
+        "title": article.title,
+        "template": template,
+        "draft_media_id": draft_media_id,
+        "preview_html": str(preview_path),
+    }
+
+    if args.publish:
+        publish_id = client.submit_publish(token, draft_media_id)
+        result["publish_id"] = publish_id
+        if args.status:
+            result["status"] = client.get_publish_status(token, publish_id)
+
+    print(json.dumps(result, ensure_ascii=False))
+
+
+if __name__ == "__main__":
+    try:
+        main()
+    except Exception as exc:
+        print(
+            json.dumps(
+                {"success": False, "error": str(exc), "type": type(exc).__name__},
+                ensure_ascii=False,
+            )
+        )
+        sys.exit(1)