22

scripts/nodejs/douyin_content_parser.js

@@ -0,0 +1,147 @@
+"use strict";
+
+/**
+ * 从混杂文本中提取抖音分享内容（链接、文案、话题）。
+ * 自 zhenqianba DouyinContentParser 移植，CommonJS 导出供 require。
+ */
+
+const DIRECT_VIDEO_PATTERN =
+  /https?:\/\/[^\s"'<>]+?\.(mp4|mov|m4v|webm|avi|mkv)(\?[^\s"'<>]*)?/i;
+
+class DouyinContentParser {
+  /**
+   * @param {string} input
+   * @returns {{ success: true, content: object } | { success: false, error: string }}
+   */
+  static parseContent(input) {
+    try {
+      const raw = String(input || "").trim();
+      if (!raw) {
+        return { success: false, error: "输入内容为空" };
+      }
+      const videoUrl = this.extractVideoUrl(raw);
+      if (!videoUrl) {
+        return { success: false, error: "未找到有效的抖音视频链接" };
+      }
+      const description = this.extractDescription(raw, videoUrl);
+      const hashtags = this.extractHashtags(raw);
+      return {
+        success: true,
+        content: {
+          videoUrl,
+          description,
+          hashtags,
+          rawText: raw,
+        },
+      };
+    } catch (error) {
+      return {
+        success: false,
+        error: `解析失败: ${error instanceof Error ? error.message : String(error)}`,
+      };
+    }
+  }
+
+  static extractVideoUrl(text) {
+    const directVideoMatch = text.match(DIRECT_VIDEO_PATTERN);
+    if (directVideoMatch && directVideoMatch[0]) {
+      return directVideoMatch[0];
+    }
+    const patterns = [
+      /https:\/\/v\.douyin\.com\/[\w_\-@]+\/?/i,
+      /https:\/\/(?:www\.)?douyin\.com\/(?:video|modal)\/\d+/i,
+      /https:\/\/iesdouyin\.com\/(?:share\/video|video)\/\d+/i,
+      /v\.douyin\.com\/[\w_\-@]+/i,
+      /douyin\.com\/(?:video|modal)\/\d+/i,
+    ];
+    for (const pattern of patterns) {
+      const match = text.match(pattern);
+      if (match) {
+        let url = match[0];
+        if (!url.startsWith("http")) {
+          url = "https://" + url;
+        }
+        return url;
+      }
+    }
+    return null;
+  }
+
+  static isDirectVideoUrl(url) {
+    return DIRECT_VIDEO_PATTERN.test(String(url || "").trim());
+  }
+
+  static extractDescription(text, videoUrl) {
+    const urlIndex = text.indexOf(videoUrl);
+    let descriptionText = "";
+    if (urlIndex > 0) {
+      descriptionText = text.substring(0, urlIndex).trim();
+    } else {
+      const descriptionMatch = text.match(/^([^h]*?)(?:https?:\/\/|$)/);
+      if (descriptionMatch && descriptionMatch[1]) {
+        descriptionText = descriptionMatch[1].trim();
+      }
+    }
+    return this.cleanDescription(descriptionText);
+  }
+
+  static cleanDescription(text) {
+    if (!text) return "";
+    let cleaned = text.replace(/^[\d\s./]+/, "").trim();
+    const startMatch = cleaned.match(/^[^\u4e00-\u9fff\w\s]*(.*)$/);
+    if (startMatch && startMatch[1]) {
+      const potential = startMatch[1].trim();
+      if (potential.length > 0) {
+        cleaned = potential;
+      }
+    }
+    if (/^[\W_]+$/.test(cleaned)) {
+      return "";
+    }
+    return cleaned;
+  }
+
+  static extractHashtags(text) {
+    const hashtagPattern = /#[\w\u4e00-\u9fff_]+/g;
+    const matches = text.match(hashtagPattern);
+    if (!matches) return [];
+    return Array.from(new Set(matches));
+  }
+
+  static extractVideoId(videoUrl) {
+    const shortMatch = videoUrl.match(/v\.douyin\.com\/([\w_\-@]+)/i);
+    if (shortMatch) return shortMatch[1];
+    const standardMatch = videoUrl.match(/\/(?:video|modal)\/(\d+)/i);
+    if (standardMatch) return standardMatch[1];
+    return null;
+  }
+
+  static generateTitleFromDescription(description, maxLength = 30) {
+    if (!description) {
+      return "分享精彩内容";
+    }
+    const sentenceEnd = description.match(/[。！？\n]/);
+    let title = "";
+    if (sentenceEnd) {
+      title = description.substring(0, sentenceEnd.index).trim();
+    } else {
+      title = description;
+    }
+    if (title.length > maxLength) {
+      title = title.substring(0, maxLength) + "...";
+    }
+    return title || "精彩分享";
+  }
+
+  static quickExtractUrl(text) {
+    const r = this.parseContent(text);
+    return r.success && r.content ? r.content.videoUrl : null;
+  }
+
+  static quickExtractDescription(text) {
+    const r = this.parseContent(text);
+    return r.success && r.content ? r.content.description : null;
+  }
+}
+
+module.exports = { DouyinContentParser, DIRECT_VIDEO_PATTERN };