@qcplay/cli 1.0.14 → 1.0.15

package/lib/wechat-article.js

@@ -1,18 +1,21 @@
-import { randomBytes } from "crypto";
+import { createHash, randomBytes } from "crypto";
 import http from "http";
 import https from "https";
 import path from "path";
 import zlib from "zlib";
 
 import * as cheerio from "cheerio";
+import { applyContentRules, matchingContentRules, matchingDownloadedImageRules } from "./content-rules.js";
 
 const WECHAT_HOST = "mp.weixin.qq.com";
-const IMAGE_UPLOAD_URL = "https://api.qingcigame.com/novel/time/literature/avatar";
+const MEDIA_UPLOAD_URL = "https://api.qingcigame.com/novel/time/literature/avatar";
 const MAX_REDIRECTS = 5;
 const MAX_HTML_BYTES = 8 * 1024 * 1024;
 const MAX_IMAGE_BYTES = 25 * 1024 * 1024;
+const MAX_VIDEO_BYTES = 512 * 1024 * 1024;
 const RICH_HTML_START = "<!-- qcplay-rich-html:start -->";
 const RICH_HTML_END = "<!-- qcplay-rich-html:end -->";
+const ARES_IGNORED_GIF_SHA256 = "138ed7ebf49170605efd81c99f516d648b43c22a972743680f6d974129e54da8";
 
 const BLOCK_TAGS = new Set([
   "article",
@@ -316,6 +319,22 @@ function assertWechatImageUrl(value, pageUrl) {
   return url;
 }
 
+function assertWechatVideoUrl(value, pageUrl) {
+  const url = new URL(value, pageUrl);
+  const hostname = url.hostname.toLowerCase();
+  const allowedHost =
+    hostname === WECHAT_HOST ||
+    hostname.endsWith(".weixin.qq.com") ||
+    hostname.endsWith(".qq.com") ||
+    hostname === "qpic.cn" ||
+    hostname.endsWith(".qpic.cn");
+  if (url.protocol !== "https:" || !allowedHost) {
+    throw new Error(`视频地址不属于受支持的微信或腾讯媒体域名: ${url.origin}`);
+  }
+  url.hash = "";
+  return url;
+}
+
 function extensionForImage(contentType, sourceUrl) {
   const normalizedType = String(contentType || "").split(";", 1)[0].trim().toLowerCase();
   const extensions = {
@@ -335,6 +354,27 @@ function extensionForImage(contentType, sourceUrl) {
   throw new Error(`不支持的图片类型: ${normalizedType || "未知"}`);
 }
 
+function extensionForVideo(contentType, sourceUrl) {
+  const normalizedType = String(contentType || "").split(";", 1)[0].trim().toLowerCase();
+  const extensions = {
+    "video/mp4": ".mp4",
+    "video/quicktime": ".mov",
+    "video/webm": ".webm",
+    "video/x-matroska": ".mkv",
+    "video/x-msvideo": ".avi",
+    "video/x-ms-wmv": ".wmv"
+  };
+  if (extensions[normalizedType]) {
+    return extensions[normalizedType];
+  }
+
+  const sourceExtension = path.extname(sourceUrl.pathname).toLowerCase();
+  if ([".mp4", ".mov", ".webm", ".mkv", ".avi", ".wmv"].includes(sourceExtension)) {
+    return sourceExtension;
+  }
+  throw new Error(`不支持的视频类型: ${normalizedType || "未知"}`);
+}
+
 export async function downloadWechatImage(sourceUrl) {
   const response = await requestBuffer(sourceUrl, {
     maxBytes: MAX_IMAGE_BYTES,
@@ -355,17 +395,50 @@ export async function downloadWechatImage(sourceUrl) {
   };
 }
 
-export async function uploadOfficialImage(image, uploadUrl = IMAGE_UPLOAD_URL) {
+export async function downloadWechatVideo(sourceUrl) {
+  const videoUrl = assertWechatVideoUrl(sourceUrl, sourceUrl);
+  const response = await requestBuffer(videoUrl, {
+    maxBytes: MAX_VIDEO_BYTES,
+    validateRedirect: redirectUrl => assertWechatVideoUrl(redirectUrl, redirectUrl),
+    headers: {
+      Accept: "video/*,*/*;q=0.8",
+      Referer: "https://mp.weixin.qq.com/"
+    }
+  });
+  const headerContentType = String(response.headers["content-type"] || "").split(";", 1)[0].trim().toLowerCase();
+  let contentType = headerContentType;
+  if (!contentType.startsWith("video/")) {
+    const extension = extensionForVideo(headerContentType, response.url);
+    const types = {
+      ".mp4": "video/mp4",
+      ".mov": "video/quicktime",
+      ".webm": "video/webm",
+      ".mkv": "video/x-matroska",
+      ".avi": "video/x-msvideo",
+      ".wmv": "video/x-ms-wmv"
+    };
+    contentType = types[extension];
+  }
+  return {
+    buffer: response.body,
+    contentType,
+    sourceUrl: response.url
+  };
+}
+
+async function uploadOfficialMedia(media, kind, uploadUrl = MEDIA_UPLOAD_URL) {
   const target = new URL(uploadUrl);
-  const extension = extensionForImage(image.contentType, image.sourceUrl);
-  const filename = `wechat-${Date.now()}-${randomBytes(5).toString("hex")}${extension}`;
+  const extension = kind === "video"
+    ? extensionForVideo(media.contentType, media.sourceUrl)
+    : extensionForImage(media.contentType, media.sourceUrl);
+  const filename = `wechat-${kind}-${Date.now()}-${randomBytes(5).toString("hex")}${extension}`;
   const boundary = `----qcplay-${randomBytes(12).toString("hex")}`;
   const header = Buffer.from(
     `--${boundary}\r\nContent-Disposition: form-data; name="file"; filename="${filename}"\r\n` +
-      `Content-Type: ${image.contentType}\r\n\r\n`
+      `Content-Type: ${media.contentType}\r\n\r\n`
   );
   const footer = Buffer.from(`\r\n--${boundary}--\r\n`);
-  const body = Buffer.concat([header, image.buffer, footer]);
+  const body = Buffer.concat([header, media.buffer, footer]);
   const response = await requestBuffer(target, {
     method: "POST",
     body,
@@ -382,30 +455,170 @@ export async function uploadOfficialImage(image, uploadUrl = IMAGE_UPLOAD_URL) {
   try {
     result = JSON.parse(response.body.toString("utf8"));
   } catch {
-    throw new Error(`图片上传接口返回了非 JSON 内容: ${response.body.toString("utf8").slice(0, 160)}`);
+    throw new Error(`${kind === "video" ? "视频" : "图片"}上传接口返回了非 JSON 内容: ${response.body.toString("utf8").slice(0, 160)}`);
   }
   if (Number(result.code) !== 200 || !result.data?.path) {
-    throw new Error(result.message || `图片上传失败，接口返回 code=${result.code ?? "未知"}`);
+    throw new Error(result.message || `${kind === "video" ? "视频" : "图片"}上传失败，接口返回 code=${result.code ?? "未知"}`);
   }
 
   let uploadedUrl;
   try {
     uploadedUrl = new URL(result.data.path);
   } catch {
-    throw new Error("图片上传接口没有返回有效的图片地址");
+    throw new Error(`${kind === "video" ? "视频" : "图片"}上传接口没有返回有效的媒体地址`);
   }
   if (!new Set(["http:", "https:"]).has(uploadedUrl.protocol)) {
-    throw new Error("图片上传接口返回了不支持的图片地址");
+    throw new Error(`${kind === "video" ? "视频" : "图片"}上传接口返回了不支持的媒体地址`);
   }
   return uploadedUrl.toString();
 }
 
+export function uploadOfficialImage(image, uploadUrl = MEDIA_UPLOAD_URL) {
+  return uploadOfficialMedia(image, "image", uploadUrl);
+}
+
+export function uploadOfficialVideo(video, uploadUrl = MEDIA_UPLOAD_URL) {
+  return uploadOfficialMedia(video, "video", uploadUrl);
+}
+
 function normalizeTextContent(value) {
   return String(value || "")
     .replace(/\u00a0/g, " ")
     .replace(/[\t\r\n ]+/g, " ");
 }
 
+function isAresWechatAuthor(value) {
+  return /^阿瑞斯病毒(?:2)?$/.test(String(value || "").trim());
+}
+
+function isIgnoredAresWechatGif(image, author) {
+  if (!isAresWechatAuthor(author) || !Buffer.isBuffer(image?.buffer)) return false;
+  const contentType = String(image.contentType || "").split(";", 1)[0].trim().toLowerCase();
+  if (contentType !== "image/gif") return false;
+  return createHash("sha256").update(image.buffer).digest("hex") === ARES_IGNORED_GIF_SHA256;
+}
+
+const WECHAT_GUIDANCE_BLOCKS = "p,li,blockquote,figcaption,h1,h2,h3,h4,h5,h6,div,section,aside,footer";
+const WECHAT_GUIDANCE_PATTERNS = [
+  /(?:请|记得|欢迎|立即|点击|长按)?(?:关注|订阅).{0,12}(?:微信公众号|公众号)/,
+  /(?:长按|扫描|扫码).{0,20}(?:二维码|识别).{0,24}(?:公众号|微信|关注)/,
+  /(?:二维码|识别).{0,20}(?:长按|扫描|扫码).{0,24}(?:公众号|微信|关注)/,
+  /(?:点击|戳).{0,12}(?:阅读原文|下方名片|公众号卡片|蓝字)/,
+  /(?:点击|点亮|点个).{0,10}(?:在看|赞)|(?:转发|分享).{0,16}(?:朋友圈|微信群|微信好友)/,
+  /(?:点击).{0,12}(?:右上角|右下角).{0,16}(?:转发|分享|在看)/,
+  /(?:设为星标|星标(?:本)?公众号)/,
+  /(?:公众号|微信).{0,16}(?:后台回复|回复关键词|菜单栏)|(?:后台回复|回复关键词).{0,20}(?:领取|获取|查看|下载)/,
+  /(?:引流)?社群|(?:加入|添加|扫码加入).{0,16}(?:官方群|交流群|QQ群|社群)/,
+  /(?:官方)?游戏福利官.{0,20}(?:好友圈|好友)|(?:扫码|添加).{0,16}(?:游戏)?福利官(?:好友)?/,
+  /(?:官网|游戏内).{0,12}充值|充值.{0,12}(?:官网|入口)/,
+  /小姬有话说/
+];
+const WECHAT_MEDIA_GUIDANCE_PATTERN = /二维码|扫码|长按识别|公众号卡片|下方名片|社群|交流群|福利官|好友圈|求关注|星标|充值/;
+
+function isWechatGuidanceText(value) {
+  const text = normalizeTextContent(value).trim();
+  return Boolean(text && text.length <= 240 && WECHAT_GUIDANCE_PATTERNS.some(pattern => pattern.test(text)));
+}
+
+function isImageOnlyBlock(element) {
+  if (!element?.length) return false;
+  if (element.is("img")) return true;
+  return !normalizeTextContent(element.text()).trim() && element.find("img").length > 0;
+}
+
+export function stripWechatGuidanceHtml(value, options = {}) {
+  const $ = cheerio.load(String(value || ""), null, false);
+  const preserveTextPatterns = Array.isArray(options.preserveTextPatterns)
+    ? options.preserveTextPatterns.filter(pattern => pattern instanceof RegExp)
+    : [];
+  const protectedBlocks = new Set();
+  const shouldPreserve = value => {
+    const text = normalizeTextContent(value).trim();
+    return Boolean(text && preserveTextPatterns.some(pattern => pattern.test(text)));
+  };
+  const preservationBlocks = $(WECHAT_GUIDANCE_BLOCKS)
+    .toArray()
+    .filter(element => shouldPreserve($(element).text()))
+    .filter(element => {
+      return !$(element)
+        .find(WECHAT_GUIDANCE_BLOCKS)
+        .toArray()
+        .some(child => shouldPreserve($(child).text()));
+    });
+  preservationBlocks.forEach(element => {
+    protectedBlocks.add(element);
+  });
+  const isProtected = element => {
+    if (!element) return false;
+    if (protectedBlocks.has(element)) return true;
+    return $(element)
+      .parents()
+      .toArray()
+      .some(parent => protectedBlocks.has(parent));
+  };
+  const matchingBlocks = $(WECHAT_GUIDANCE_BLOCKS)
+    .toArray()
+    .filter(element => !isProtected(element))
+    .filter(element => isWechatGuidanceText($(element).text()))
+    .filter(element => {
+      return !$(element)
+        .find(WECHAT_GUIDANCE_BLOCKS)
+        .toArray()
+        .some(child => isWechatGuidanceText($(child).text()));
+    });
+
+  for (const element of matchingBlocks) {
+    const block = $(element);
+    const text = normalizeTextContent(block.text()).trim();
+    let target = block;
+    if (/社群|官方群|交流群|QQ群|福利官|好友圈/.test(text)) {
+      for (const parentNode of block.parents("section,div").toArray()) {
+        const parent = $(parentNode);
+        const parentText = normalizeTextContent(parent.text()).trim();
+        const hasAssociatedImage =
+          isImageOnlyBlock(parent.prev()) || isImageOnlyBlock(parent.next()) || parent.find("img").length > 0;
+        if (parentText.length <= 500 && parent.find(WECHAT_GUIDANCE_BLOCKS).length <= 12 && hasAssociatedImage) {
+          target = parent;
+          break;
+        }
+      }
+    }
+    if (WECHAT_MEDIA_GUIDANCE_PATTERN.test(text)) {
+      const previous = target.prev();
+      const next = target.next();
+      if (isImageOnlyBlock(previous) && !isProtected(previous.get(0))) previous.remove();
+      if (isImageOnlyBlock(next) && !isProtected(next.get(0))) next.remove();
+    }
+    target.remove();
+  }
+
+  $("img").each((_, element) => {
+    if (isProtected(element)) return;
+    const image = $(element);
+    const description = [image.attr("alt"), image.attr("title"), image.attr("aria-label")]
+      .filter(Boolean)
+      .join(" ");
+    if (!/(?:公众号)?二维码|扫码关注|长按识别|引流社群|求关注|星标|官网充值/.test(description)) return;
+    const container = image.closest("p,figure,div,section").first();
+    if (container.length && isImageOnlyBlock(container)) {
+      container.remove();
+    } else {
+      image.remove();
+    }
+  });
+
+  $(WECHAT_GUIDANCE_BLOCKS)
+    .toArray()
+    .reverse()
+    .forEach(element => {
+      const block = $(element);
+      if (!normalizeTextContent(block.text()).trim() && block.find("img,table,hr").length === 0) {
+        block.remove();
+      }
+    });
+  return $.root().html().trim();
+}
+
 function escapeMarkdownText(value) {
   return normalizeTextContent(value).replace(/([\\`*_\[\]])/g, "\\$1");
 }
@@ -464,7 +677,16 @@ function elementTextColor(element) {
 }
 
 function safeContentUrl(element) {
-  const attributes = ["href", "src", "data-src", "data-url", "data-link", "data-video-url", "data-audio-url"];
+  const attributes = [
+    "data-qcplay-media-src",
+    "href",
+    "src",
+    "data-src",
+    "data-url",
+    "data-link",
+    "data-video-url",
+    "data-audio-url"
+  ];
   const candidates = [];
   for (const attribute of attributes) {
     candidates.push(element.attr(attribute));
@@ -531,6 +753,18 @@ function safeHttpUrl(value) {
   }
 }
 
+function directWechatVideoUrl(value) {
+  const candidate = safeHttpUrl(value);
+  if (!candidate) return "";
+  const url = new URL(candidate);
+  const hostname = url.hostname.toLowerCase();
+  const extension = path.extname(url.pathname).toLowerCase();
+  if ([".mp4", ".mov", ".webm", ".mkv", ".avi", ".wmv"].includes(extension)) {
+    return candidate;
+  }
+  return hostname === "mpvideo.qpic.cn" || hostname.endsWith(".mpvideo.qpic.cn") ? candidate : "";
+}
+
 function sanitizeInlineStyle(value, backgroundUrls = []) {
   const declarations = [];
   for (const declaration of String(value || "").split(";")) {
@@ -602,6 +836,54 @@ function isStandaloneImage(element) {
     });
 }
 
+function removeWechatImageLoadingPlaceholders($, content) {
+  const loadingClassNames = new Set([
+    "js_img_loading",
+    "wx_img_loading",
+    "wx_img_loading_msg",
+    "img_loading",
+    "img_loading_msg"
+  ]);
+  const hasLoadingClass = element =>
+    String(element.attr("class") || "")
+      .split(/\s+/)
+      .map(className => className.toLowerCase())
+      .some(className => loadingClassNames.has(className));
+
+  // WeChat can wrap the real image in a loading node. Remove loader-only
+  // descendants, but never remove an ancestor that owns an image.
+  const candidates = content
+    .find("*")
+    .toArray()
+    .filter(node => hasLoadingClass($(node)))
+    .sort((left, right) => $(right).parents().length - $(left).parents().length);
+  for (const node of candidates) {
+    const element = $(node);
+    const tag = String(node.tagName || node.name || "").toLowerCase();
+    if (tag === "img") continue;
+    if (element.find("img").length === 0) {
+      element.remove();
+      continue;
+    }
+    element.find("*").each((_, child) => {
+      const childElement = $(child);
+      if (hasLoadingClass(childElement) && childElement.find("img").length === 0) {
+        childElement.remove();
+      }
+    });
+    const remainingClasses = String(element.attr("class") || "")
+      .split(/\s+/)
+      .filter(className => className && !loadingClassNames.has(className.toLowerCase()));
+    if (remainingClasses.length > 0) element.attr("class", remainingClasses.join(" "));
+    else element.removeAttr("class");
+  }
+}
+
+function meaningfulImageAlt(value) {
+  const alt = normalizeTextContent(value).trim();
+  return /^(?:图片|图像|image|picture)$/i.test(alt) ? "" : alt;
+}
+
 export function sanitizeArticleRichHtml(value, options = {}) {
   const initial = cheerio.load(String(value || ""), null, false);
   const existingRoot = initial('[data-qcplay-rich-root="1"]').first();
@@ -641,8 +923,16 @@ export function sanitizeArticleRichHtml(value, options = {}) {
     const transferredBackgrounds = String(originalAttributes["data-qcplay-background-srcs"] || "")
       .split("|")
       .filter(Boolean);
-    const backgroundUrls = transferredBackgrounds.length > 0 ? transferredBackgrounds : backgroundImageSources(element);
-    const sanitizedStyle = sanitizeInlineStyle(originalAttributes.style || originalAttributes["data-style"], backgroundUrls);
+    const backgroundUrls = Object.prototype.hasOwnProperty.call(originalAttributes, "data-qcplay-background-processed")
+      ? transferredBackgrounds
+      : transferredBackgrounds.length > 0
+        ? transferredBackgrounds
+        : backgroundImageSources(element);
+    let sanitizedStyle = sanitizeInlineStyle(originalAttributes.style || originalAttributes["data-style"], backgroundUrls);
+    const attributeColor = normalizeArticleColor(originalAttributes.color || originalAttributes["data-color"]);
+    if (tag !== "font" && attributeColor && !/(?:^|;)\s*(?:color|-webkit-text-fill-color)\s*:/i.test(sanitizedStyle)) {
+      sanitizedStyle = [sanitizedStyle, `color:${attributeColor}`].filter(Boolean).join(";");
+    }
     for (const attribute of Object.keys(originalAttributes)) {
       element.removeAttr(attribute);
     }
@@ -659,7 +949,7 @@ export function sanitizeArticleRichHtml(value, options = {}) {
         element.parent().attr("style") || ""
       );
       const src = safeHttpUrl(originalAttributes["data-qcplay-src"] || originalAttributes.src);
-      const alt = normalizeTextContent(originalAttributes.alt || "图片").trim();
+      const alt = meaningfulImageAlt(originalAttributes.alt);
       if (!src) {
         element.replaceWith(alt ? $("<span></span>").text(alt) : "");
         continue;
@@ -722,8 +1012,8 @@ export function sanitizeArticleRichHtml(value, options = {}) {
       continue;
     }
 
-    if (tag === "font" && normalizeArticleColor(originalAttributes.color)) {
-      element.attr("color", normalizeArticleColor(originalAttributes.color));
+    if (tag === "font" && attributeColor) {
+      element.attr("color", attributeColor);
     }
     if ((tag === "td" || tag === "th") && /^\d+$/.test(originalAttributes.colspan || "")) {
       element.attr("colspan", originalAttributes.colspan);
@@ -860,7 +1150,7 @@ function markdownForNode($, node, listDepth = 0, inheritedColor = "", context) {
       const description = elementDescription(element);
       return description ? `\n\n${escapeMarkdownText(description)}\n\n` : "";
     }
-    const alt = escapeMarkdownText(element.attr("alt") || "图片");
+    const alt = escapeMarkdownText(meaningfulImageAlt(element.attr("alt")));
     return `\n\n![${alt}](${src})\n\n`;
   }
   if (tag === "strong" || tag === "b") {
@@ -999,8 +1289,31 @@ export async function convertWechatHtml(html, pageUrl, options = {}) {
   if (!title) {
     throw new Error("微信页面中没有找到文章标题");
   }
+  const author = $("#js_name").first().text().trim() || $('meta[name="author"]').attr("content")?.trim() || "";
 
   content.find("script,style").remove();
+  removeWechatImageLoadingPlaceholders($, content);
+  const ruleArticle = { html: content.html(), meta: { source_url: sourceUrl.toString() } };
+  const ruleEntry = {
+    project: options.project || (isAresWechatAuthor(author) ? "阿瑞斯病毒2" : ""),
+    region: options.region || "",
+    platform: options.platform || "*",
+    platformKey: options.platform || "*"
+  };
+  const matchedContentRules = options.contentRules
+    ? matchingContentRules(options.contentRules, ruleArticle, ruleEntry, { phase: "import" }).map(rule => rule.name)
+    : [];
+  const appliedContentRules = new Set();
+  if (options.contentRules) {
+    const rulePrepared = applyContentRules(
+      ruleArticle,
+      ruleEntry,
+      options.contentRules,
+      { phase: "import" }
+    );
+    if (rulePrepared.html !== content.html()) content.html(rulePrepared.html);
+    rulePrepared.appliedContentRules?.forEach(name => appliedContentRules.add(name));
+  }
   const sourceTextNodes = collectMeaningfulTextNodes(content.get(0));
   const imageElements = content.find("img").toArray();
   const sourceImages = imageElements.map(element => {
@@ -1013,22 +1326,42 @@ export async function convertWechatHtml(html, pageUrl, options = {}) {
     .map(element => ({ element, sources: backgroundImageSources($(element)) }))
     .filter(entry => entry.sources.length > 0);
   const backgroundSources = backgroundEntries.flatMap(entry => entry.sources);
+  const videoElements = content.find("video,mp-common-videosnap").toArray();
+  const videoSources = videoElements.map(element => directWechatVideoUrl(safeContentUrl($(element))));
   const coverSource = $('meta[property="og:image"]').attr("content") || "";
   const allSources = [coverSource, ...sourceImages, ...backgroundSources].filter(Boolean);
   const uploaded = new Map();
+  const ignored = new Set();
   const downloadImage = options.downloadImage || downloadWechatImage;
   const uploadImage = options.uploadImage || uploadOfficialImage;
+  const downloadVideo = options.downloadVideo || downloadWechatVideo;
+  const uploadVideo = options.uploadVideo || uploadOfficialVideo;
+  const uniqueSourceCount = new Set(allSources.map(rawSource => assertWechatImageUrl(rawSource, sourceUrl).toString())).size;
+  let processedSourceCount = 0;
 
   for (const rawSource of allSources) {
     const imageUrl = assertWechatImageUrl(rawSource, sourceUrl);
     const key = imageUrl.toString();
-    if (uploaded.has(key)) {
+    if (uploaded.has(key) || ignored.has(key)) {
       continue;
     }
-    const current = uploaded.size + 1;
-    options.onProgress?.({ current, total: new Set(allSources).size, sourceUrl: key });
+    const current = ++processedSourceCount;
+    options.onProgress?.({ mediaType: "image", current, total: uniqueSourceCount, sourceUrl: key });
     try {
       const image = await downloadImage(imageUrl);
+      const matchedImageRules = options.contentRules
+        ? matchingDownloadedImageRules(image, ruleArticle, ruleEntry, options.contentRules, { phase: "import" })
+        : [];
+      const shouldIgnore =
+        matchedImageRules.length > 0 ||
+        (options.shouldIgnoreImage
+          ? await options.shouldIgnoreImage(image, { author, sourceUrl: key })
+          : isIgnoredAresWechatGif(image, author));
+      if (shouldIgnore) {
+        matchedImageRules.forEach(rule => appliedContentRules.add(rule.name));
+        ignored.add(key);
+        continue;
+      }
       uploaded.set(key, await uploadImage(image));
     } catch (error) {
       throw new Error(`第 ${current} 张图片处理失败: ${error.message}`);
@@ -1041,6 +1374,10 @@ export async function convertWechatHtml(html, pageUrl, options = {}) {
       return;
     }
     const key = assertWechatImageUrl(rawSource, sourceUrl).toString();
+    if (ignored.has(key)) {
+      $(element).remove();
+      return;
+    }
     $(element).attr("data-qcplay-src", uploaded.get(key));
   });
 
@@ -1049,9 +1386,30 @@ export async function convertWechatHtml(html, pageUrl, options = {}) {
       const key = assertWechatImageUrl(rawSource, sourceUrl).toString();
       return uploaded.get(key);
     });
+    $(entry.element).attr("data-qcplay-background-processed", "1");
     $(entry.element).attr("data-qcplay-background-srcs", urls.filter(Boolean).join("|"));
   });
 
+  const uploadedVideos = new Map();
+  const uniqueVideoSources = [...new Set(videoSources.filter(Boolean))];
+  for (const rawSource of uniqueVideoSources) {
+    const current = uploadedVideos.size + 1;
+    options.onProgress?.({ mediaType: "video", current, total: uniqueVideoSources.length, sourceUrl: rawSource });
+    try {
+      const video = await downloadVideo(rawSource);
+      uploadedVideos.set(rawSource, await uploadVideo(video));
+    } catch (error) {
+      throw new Error(`第 ${current} 个视频处理失败: ${error.message}`);
+    }
+  }
+
+  videoElements.forEach((element, index) => {
+    const source = videoSources[index];
+    if (source) {
+      $(element).attr("data-qcplay-media-src", uploadedVideos.get(source));
+    }
+  });
+
   const richHtml = sanitizeArticleRichHtml(content.html());
   if (!richHtml) {
     throw new Error("微信文章富文本转换后为空");
@@ -1069,7 +1427,9 @@ export async function convertWechatHtml(html, pageUrl, options = {}) {
       .join(" / ");
     throw new Error(`微信文章有 ${missingRichTextNodes.length} 处正文未保留原始格式: ${preview}`);
   }
-  const sourceImageCount = sourceImages.filter(Boolean).length;
+  const sourceImageCount = sourceImages.filter(rawSource => {
+    return rawSource && !ignored.has(assertWechatImageUrl(rawSource, sourceUrl).toString());
+  }).length;
   const richImageCount = richDocument("img").length;
   if (richImageCount !== sourceImageCount) {
     throw new Error(`微信文章图片结构不完整: 原文 ${sourceImageCount} 张，转换后 ${richImageCount} 张`);
@@ -1104,7 +1464,7 @@ export async function convertWechatHtml(html, pageUrl, options = {}) {
 
   return {
     title,
-    author: $("#js_name").first().text().trim() || $('meta[name="author"]').attr("content")?.trim() || "",
+    author,
     releaseDate: releaseDateFromWechatPage($, html),
     excerpt: $('meta[property="og:description"]').attr("content")?.trim() || "",
     thumbnail: coverSource ? uploaded.get(assertWechatImageUrl(coverSource, sourceUrl).toString()) || "" : uploaded.values().next().value || "",
@@ -1112,6 +1472,7 @@ export async function convertWechatHtml(html, pageUrl, options = {}) {
     richHtml,
     richImageCount,
     imageCount: uploaded.size,
+    videoCount: uploadedVideos.size,
     backgroundImageCount: backgroundSources.length,
     colorCount: markdownContext.colorCount,
     colors: [...markdownContext.colors],
@@ -1119,7 +1480,9 @@ export async function convertWechatHtml(html, pageUrl, options = {}) {
     specialElementCount: markdownContext.specialElementCount,
     tableCount: markdownContext.tableCount,
     textSegmentCount: sourceTextNodes.length,
-    sourceUrl: sourceUrl.toString()
+    sourceUrl: sourceUrl.toString(),
+    matchedContentRules,
+    appliedContentRules: [...appliedContentRules]
   };
 }
 
@@ -1127,6 +1490,17 @@ function yamlString(value) {
   return JSON.stringify(String(value || ""));
 }
 
+function distributionProjectForWechatArticle(article) {
+  const author = String(article.author || "").trim();
+  if (isAresWechatAuthor(author)) return "阿瑞斯病毒2";
+  return "";
+}
+
+function gameIdForWechatArticle(article) {
+  const author = String(article.author || "").trim();
+  return isAresWechatAuthor(author) ? "41" : "39";
+}
+
 export function buildOfficialArticleMarkdown(article) {
   const articleBody = article.richHtml
     ? `${RICH_HTML_START}\n${article.richHtml}\n${RICH_HTML_END}`
@@ -1146,10 +1520,11 @@ is_index: "0"
 release_time: ${yamlString(article.releaseDate)}
 area: "1"
 sort: "100"
-game_id: "39"
+game_id: ${yamlString(article.gameId || gameIdForWechatArticle(article))}
 index_pc_img: ""
 index_move_img: ""
 source_url: ${yamlString(article.sourceUrl)}
+distribution_project: ${yamlString(article.distributionProject || distributionProjectForWechatArticle(article))}
 ---
 
 ${articleBody}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@qcplay/cli",
-  "version": "1.0.14",
+  "version": "1.0.15",
   "description": "QCPlay CLI",
   "type": "module",
   "bin": {
@@ -29,6 +29,9 @@
     "test:wechat": "node test/wechat-article.test.js",
     "test:lark": "node test/lark-article.test.js",
     "test:platforms": "node test/platform-publish.test.js",
+    "test:rules": "node test/content-rules.test.js",
+    "test:rules-cli": "node test/content-rules-cli.test.js",
+    "test:project-catalog": "node test/project-catalog.test.js",
     "build:cli": "node --check bin/qcplay.js",
     "pack:check": "npm pack --dry-run"
   },