koztv-blog-tools 1.1.1 → 1.2.1

package/README.md

@@ -0,0 +1,215 @@
+# koztv-blog-tools
+
+Shared utilities for Telegram-based blog sites. Export posts from Telegram channels, translate with LLM APIs, and generate markdown for static site generators.
+
+## Installation
+
+```bash
+npm install koztv-blog-tools
+```
+
+## Features
+
+- **Telegram Export** — Export full channel history via MTProto (gramjs)
+- **Translation** — Translate posts using OpenAI-compatible APIs (GLM, OpenAI, etc.)
+- **Multi-language** — Export to multiple languages simultaneously
+- **Markdown Generation** — Create markdown files with YAML frontmatter
+- **Media Download** — Download photos, videos, and documents
+- **Incremental Export** — Track processed posts, skip duplicates
+
+## Quick Start
+
+### Export + Translate
+
+```javascript
+const { exportAndTranslate } = require('koztv-blog-tools');
+
+const result = await exportAndTranslate({
+  // Telegram credentials (from https://my.telegram.org)
+  apiId: 12345678,
+  apiHash: 'your_api_hash',
+  session: 'saved_session_string', // from previous auth
+
+  // Target channel
+  channel: '@channelname',
+
+  // Output
+  outputDir: './content/posts',
+  mediaDir: './public/media',
+
+  // Translation (optional)
+  translate: {
+    apiKey: 'your_llm_api_key',
+    apiUrl: 'https://api.openai.com/v1', // or GLM, etc.
+    model: 'gpt-4',
+    sourceLang: 'ru',
+    targetLangs: ['en', 'de'],  // translate to multiple languages
+    keepOriginal: true,         // also save original language
+  },
+
+  // Callbacks
+  onProgress: (msg) => console.log(msg),
+  onSession: (s) => saveSession(s), // save for future use
+});
+
+console.log(`Exported: ${result.exported}, Processed: ${result.processed}`);
+```
+
+### Export Only (no translation)
+
+```javascript
+const { exportTelegramChannel } = require('koztv-blog-tools');
+
+const result = await exportTelegramChannel({
+  apiId: 12345678,
+  apiHash: 'your_api_hash',
+  session: 'saved_session_string',
+  target: '@channelname',
+  outputDir: './export',
+  downloadMedia: true,
+  limit: 100, // optional: limit number of posts
+  since: new Date('2024-01-01'), // optional: filter by date
+});
+```
+
+### Translate Text
+
+```javascript
+const { translateContent, translateTitle } = require('koztv-blog-tools');
+
+const translated = await translateContent('Привет мир', {
+  apiKey: 'your_api_key',
+  apiUrl: 'https://api.openai.com/v1',
+  model: 'gpt-4',
+  sourceLang: 'ru',
+  targetLang: 'en',
+});
+// => "Hello world"
+```
+
+## Authentication
+
+First-time authentication requires QR code login. See `scripts/qr-login.js` in your project:
+
+```javascript
+// Example QR login script
+const { TelegramClient } = require('telegram');
+const { StringSession } = require('telegram/sessions');
+
+const client = new TelegramClient(
+  new StringSession(''),
+  API_ID,
+  API_HASH,
+  { connectionRetries: 5 }
+);
+
+await client.start({
+  phoneNumber: async () => prompt('Phone: '),
+  password: async () => prompt('2FA: '),
+  phoneCode: async () => prompt('Code: '),
+  onError: console.error,
+});
+
+console.log('Session:', client.session.save());
+```
+
+## Output Structure
+
+With multi-language enabled (`targetLangs` + `keepOriginal`):
+
+```
+content/posts/
+  en/
+    my-post-slug/
+      index.md
+  ru/
+    my-post-slug/
+      index.md
+public/media/
+  000123/
+    image1.jpg
+    image2.jpg
+```
+
+Markdown format:
+
+```yaml
+---
+title: "Post Title"
+date: 2024-01-15
+lang: en
+original_link: "https://t.me/channel/123"
+translated_from: "ru"
+---
+
+Post content here...
+
+![](/media/000123/image1.jpg)
+```
+
+## Environment Variables
+
+For GitHub Actions / CI:
+
+```bash
+TELEGRAM_API_ID=12345678
+TELEGRAM_API_HASH=your_api_hash
+TELEGRAM_SESSION=base64_session_string
+TELEGRAM_CHANNEL=@channelname
+
+LLM_API_KEY=your_llm_key
+LLM_API_URL=https://api.openai.com/v1
+LLM_MODEL=gpt-4
+
+TARGET_LANGS=en,de      # comma-separated
+KEEP_ORIGINAL=true      # keep source language
+```
+
+## API Reference
+
+### exportAndTranslate(options)
+
+Main function for export + translation workflow.
+
+**Options:**
+- `apiId`, `apiHash`, `session` — Telegram credentials
+- `channel` — Target channel (@username or ID)
+- `outputDir` — Where to save markdown files
+- `mediaDir` — Where to save media (optional)
+- `limit` — Max posts to export (optional)
+- `since` — Export posts after this date (optional)
+- `downloadMedia` — Download media files (default: true)
+- `translate` — Translation config (optional)
+- `onProgress` — Progress callback
+- `onSession` — Session save callback
+- `processedLog` — Object tracking processed posts
+- `onProcessedLog` — Callback to save processed log
+
+### exportTelegramChannel(options)
+
+Low-level Telegram export.
+
+### translateContent(text, options)
+
+Translate text content.
+
+### translateTitle(title, options)
+
+Translate title (optimized prompt for short text).
+
+### generateEnglishSlug(title, options)
+
+Generate URL-friendly slug from any language title.
+
+## Used By
+
+- [koz.tv](https://koz.tv) — Personal blog with Telegram sync
+- [staskoz.com](https://staskoz.com) — Another blog using this package
+
+## Related
+
+- [k-engine](https://github.com/Koz-TV/k-engine) — Static site generator with multi-language support
+
+## License
+
+MIT

package/dist/index.d.mts

@@ -216,4 +216,72 @@ declare function resumeExport(options: Omit<TelegramExportOptions, 'onPhoneNumbe
     session: string;
 }): Promise<ExportResult>;
 
-export { type AnalyticsConfig, type ExportResult, type ExportedPost, type GoalName, type GoalParams, type GroupedPost, type ParsePostOptions, type Post, type TelegramExportOptions, type TranslateOptions, categorizePost, cleanContent, configureAnalytics, deduplicatePosts, exportTelegramChannel, extractAttachments, extractExcerpt, extractTitle, formatPostMarkdown, generateEnglishSlug, generateSlug, groupPosts, parsePost, resumeExport, trackBookAppointment, trackGoal, trackLearnMore, trackServiceClick, trackTelegramClick, translateContent, translateTitle };
+/**
+ * Combined Telegram export + translation module
+ */
+interface TranslationConfig {
+    /** API key for translation service */
+    apiKey: string;
+    /** API base URL (OpenAI-compatible) */
+    apiUrl: string;
+    /** Model name */
+    model: string;
+    /** Source language code (default: 'ru') */
+    sourceLang?: string;
+    /** Target languages array (e.g., ['en', 'de', 'zh']) */
+    targetLangs: string[];
+    /** Also keep original language version */
+    keepOriginal?: boolean;
+}
+interface ExportAndTranslateOptions {
+    /** Telegram API ID */
+    apiId: number;
+    /** Telegram API Hash */
+    apiHash: string;
+    /** Session string */
+    session?: string;
+    /** Target channel */
+    channel: string;
+    /** Output directory for posts */
+    outputDir: string;
+    /** Media output directory (default: outputDir/../media) */
+    mediaDir?: string;
+    /** Export limit */
+    limit?: number;
+    /** Export posts since date */
+    since?: Date;
+    /** Download media files */
+    downloadMedia?: boolean;
+    /** Translation config (optional - if not provided, no translation) */
+    translate?: TranslationConfig;
+    /** Progress callback */
+    onProgress?: (message: string) => void;
+    /** Session save callback */
+    onSession?: (session: string) => void;
+    /** Log of already processed posts (to skip) */
+    processedLog?: Record<string, any>;
+    /** Callback to save processed log */
+    onProcessedLog?: (log: Record<string, any>) => void;
+}
+interface ExportAndTranslateResult {
+    /** Number of posts exported */
+    exported: number;
+    /** Number of posts processed (translated/saved) */
+    processed: number;
+    /** Number of posts skipped */
+    skipped: number;
+    /** Channel metadata */
+    channelMeta: {
+        id: number;
+        username: string;
+        title: string;
+    };
+    /** Session string for future use */
+    session: string;
+}
+/**
+ * Export posts from Telegram channel with optional translation
+ */
+declare function exportAndTranslate(options: ExportAndTranslateOptions): Promise<ExportAndTranslateResult>;
+
+export { type AnalyticsConfig, type ExportAndTranslateOptions, type ExportAndTranslateResult, type ExportResult, type ExportedPost, type GoalName, type GoalParams, type GroupedPost, type ParsePostOptions, type Post, type TelegramExportOptions, type TranslateOptions, type TranslationConfig, categorizePost, cleanContent, configureAnalytics, deduplicatePosts, exportAndTranslate, exportTelegramChannel, extractAttachments, extractExcerpt, extractTitle, formatPostMarkdown, generateEnglishSlug, generateSlug, groupPosts, parsePost, resumeExport, trackBookAppointment, trackGoal, trackLearnMore, trackServiceClick, trackTelegramClick, translateContent, translateTitle };

package/dist/index.d.ts

@@ -216,4 +216,72 @@ declare function resumeExport(options: Omit<TelegramExportOptions, 'onPhoneNumbe
     session: string;
 }): Promise<ExportResult>;
 
-export { type AnalyticsConfig, type ExportResult, type ExportedPost, type GoalName, type GoalParams, type GroupedPost, type ParsePostOptions, type Post, type TelegramExportOptions, type TranslateOptions, categorizePost, cleanContent, configureAnalytics, deduplicatePosts, exportTelegramChannel, extractAttachments, extractExcerpt, extractTitle, formatPostMarkdown, generateEnglishSlug, generateSlug, groupPosts, parsePost, resumeExport, trackBookAppointment, trackGoal, trackLearnMore, trackServiceClick, trackTelegramClick, translateContent, translateTitle };
+/**
+ * Combined Telegram export + translation module
+ */
+interface TranslationConfig {
+    /** API key for translation service */
+    apiKey: string;
+    /** API base URL (OpenAI-compatible) */
+    apiUrl: string;
+    /** Model name */
+    model: string;
+    /** Source language code (default: 'ru') */
+    sourceLang?: string;
+    /** Target languages array (e.g., ['en', 'de', 'zh']) */
+    targetLangs: string[];
+    /** Also keep original language version */
+    keepOriginal?: boolean;
+}
+interface ExportAndTranslateOptions {
+    /** Telegram API ID */
+    apiId: number;
+    /** Telegram API Hash */
+    apiHash: string;
+    /** Session string */
+    session?: string;
+    /** Target channel */
+    channel: string;
+    /** Output directory for posts */
+    outputDir: string;
+    /** Media output directory (default: outputDir/../media) */
+    mediaDir?: string;
+    /** Export limit */
+    limit?: number;
+    /** Export posts since date */
+    since?: Date;
+    /** Download media files */
+    downloadMedia?: boolean;
+    /** Translation config (optional - if not provided, no translation) */
+    translate?: TranslationConfig;
+    /** Progress callback */
+    onProgress?: (message: string) => void;
+    /** Session save callback */
+    onSession?: (session: string) => void;
+    /** Log of already processed posts (to skip) */
+    processedLog?: Record<string, any>;
+    /** Callback to save processed log */
+    onProcessedLog?: (log: Record<string, any>) => void;
+}
+interface ExportAndTranslateResult {
+    /** Number of posts exported */
+    exported: number;
+    /** Number of posts processed (translated/saved) */
+    processed: number;
+    /** Number of posts skipped */
+    skipped: number;
+    /** Channel metadata */
+    channelMeta: {
+        id: number;
+        username: string;
+        title: string;
+    };
+    /** Session string for future use */
+    session: string;
+}
+/**
+ * Export posts from Telegram channel with optional translation
+ */
+declare function exportAndTranslate(options: ExportAndTranslateOptions): Promise<ExportAndTranslateResult>;
+
+export { type AnalyticsConfig, type ExportAndTranslateOptions, type ExportAndTranslateResult, type ExportResult, type ExportedPost, type GoalName, type GoalParams, type GroupedPost, type ParsePostOptions, type Post, type TelegramExportOptions, type TranslateOptions, type TranslationConfig, categorizePost, cleanContent, configureAnalytics, deduplicatePosts, exportAndTranslate, exportTelegramChannel, extractAttachments, extractExcerpt, extractTitle, formatPostMarkdown, generateEnglishSlug, generateSlug, groupPosts, parsePost, resumeExport, trackBookAppointment, trackGoal, trackLearnMore, trackServiceClick, trackTelegramClick, translateContent, translateTitle };

package/dist/index.js

@@ -34,6 +34,7 @@ __export(index_exports, {
   cleanContent: () => cleanContent,
   configureAnalytics: () => configureAnalytics,
   deduplicatePosts: () => deduplicatePosts,
+  exportAndTranslate: () => exportAndTranslate,
   exportTelegramChannel: () => exportTelegramChannel,
   extractAttachments: () => extractAttachments,
   extractExcerpt: () => extractExcerpt,
@@ -586,12 +587,253 @@ async function resumeExport(options) {
   }
   return exportTelegramChannel(options);
 }
+
+// src/exporter.ts
+var fs2 = __toESM(require("fs"));
+var path2 = __toESM(require("path"));
+var translitMap2 = {
+  "\u0430": "a",
+  "\u0431": "b",
+  "\u0432": "v",
+  "\u0433": "g",
+  "\u0434": "d",
+  "\u0435": "e",
+  "\u0451": "yo",
+  "\u0436": "zh",
+  "\u0437": "z",
+  "\u0438": "i",
+  "\u0439": "y",
+  "\u043A": "k",
+  "\u043B": "l",
+  "\u043C": "m",
+  "\u043D": "n",
+  "\u043E": "o",
+  "\u043F": "p",
+  "\u0440": "r",
+  "\u0441": "s",
+  "\u0442": "t",
+  "\u0443": "u",
+  "\u0444": "f",
+  "\u0445": "h",
+  "\u0446": "ts",
+  "\u0447": "ch",
+  "\u0448": "sh",
+  "\u0449": "sch",
+  "\u044A": "",
+  "\u044B": "y",
+  "\u044C": "",
+  "\u044D": "e",
+  "\u044E": "yu",
+  "\u044F": "ya"
+};
+function transliterate(text) {
+  return text.toLowerCase().split("").map((char) => translitMap2[char] || char).join("");
+}
+function generateSlug2(text, lang) {
+  let processed = text;
+  if (/[а-яё]/i.test(text)) {
+    processed = transliterate(text);
+  }
+  return processed.toLowerCase().replace(/[^a-z0-9\s-]/g, "").replace(/\s+/g, "-").replace(/-+/g, "-").replace(/^-|-$/g, "").substring(0, 60) || "untitled";
+}
+function cleanContent2(text) {
+  return text.replace(/##\s*Attachments\n(?:- [^\n]+\n?)*/g, "").replace(/- media\/\d+\/[^\n]+/g, "").replace(/#\w+@\w+/g, "").trim();
+}
+function extractTitleAndBody(content) {
+  const lines = content.split("\n").filter((l) => l.trim());
+  const title = lines[0]?.replace(/[#@[\]*]/g, "").trim() || "Untitled";
+  const body = lines.slice(1).join("\n").trim();
+  return { title, body };
+}
+async function processPost(post, options, exportDir) {
+  const { translate, outputDir, mediaDir, onProgress } = options;
+  const createdPosts = [];
+  const cleanedContent = cleanContent2(post.content || "");
+  const { title: originalTitle, body: originalBody } = extractTitleAndBody(cleanedContent);
+  const date = post.date.toISOString().split("T")[0];
+  const languages = [];
+  if (translate && translate.targetLangs.length > 0) {
+    const sourceLang = translate.sourceLang || "ru";
+    if (translate.keepOriginal) {
+      languages.push({
+        lang: sourceLang,
+        title: originalTitle,
+        body: originalBody,
+        isOriginal: true
+      });
+    }
+    for (const targetLang of translate.targetLangs) {
+      if (targetLang === sourceLang) continue;
+      onProgress?.(`  Translating to ${targetLang}...`);
+      const translateOpts = {
+        apiKey: translate.apiKey,
+        apiUrl: translate.apiUrl,
+        model: translate.model,
+        sourceLang,
+        targetLang
+      };
+      try {
+        const translatedTitle = await translateTitle(originalTitle, translateOpts);
+        const translatedBody = originalBody ? await translateContent(originalBody, translateOpts) : "";
+        languages.push({
+          lang: targetLang,
+          title: translatedTitle,
+          body: translatedBody,
+          isOriginal: false
+        });
+        onProgress?.(`  \u2192 ${targetLang}: "${translatedTitle.substring(0, 40)}..."`);
+      } catch (error) {
+        onProgress?.(`  Error translating to ${targetLang}: ${error.message}`);
+      }
+      await new Promise((r) => setTimeout(r, 300));
+    }
+  } else {
+    languages.push({
+      lang: "original",
+      title: originalTitle,
+      body: originalBody,
+      isOriginal: true
+    });
+  }
+  for (const { lang, title, body, isOriginal } of languages) {
+    const slug = generateSlug2(title, lang);
+    let postDir;
+    if (languages.length > 1 || translate && translate.targetLangs.length > 0) {
+      postDir = path2.join(outputDir, lang, slug);
+    } else {
+      postDir = path2.join(outputDir, slug);
+    }
+    if (fs2.existsSync(postDir)) {
+      onProgress?.(`  Skipping existing: ${lang}/${slug}`);
+      continue;
+    }
+    fs2.mkdirSync(postDir, { recursive: true });
+    let finalBody = body;
+    const mediaFiles = [];
+    if (post.mediaFiles && post.mediaFiles.length > 0) {
+      const paddedId = String(post.msgId).padStart(6, "0");
+      const targetMediaDir = mediaDir ? path2.join(mediaDir, paddedId) : path2.join(outputDir, "..", "media", paddedId);
+      fs2.mkdirSync(targetMediaDir, { recursive: true });
+      for (let i = 0; i < post.mediaFiles.length; i++) {
+        const sourcePath = path2.join(exportDir, post.mediaFiles[i]);
+        const ext = path2.extname(sourcePath).toLowerCase() || ".jpg";
+        const newName = `image${i + 1}${ext}`;
+        const targetPath = path2.join(targetMediaDir, newName);
+        if (fs2.existsSync(sourcePath) && !fs2.existsSync(targetPath)) {
+          fs2.copyFileSync(sourcePath, targetPath);
+          mediaFiles.push(newName);
+        }
+      }
+      if (mediaFiles.length > 0) {
+        const images = mediaFiles.filter((f) => !f.match(/\.(mp4|mov|webm|m4v)$/i));
+        const videos = mediaFiles.filter((f) => f.match(/\.(mp4|mov|webm|m4v)$/i));
+        const imageMarkdown = images.map((f) => `![](/media/${paddedId}/${f})`).join("\n\n");
+        const videoMarkdown = videos.map((f) => `<video src="/media/${paddedId}/${f}" controls></video>`).join("\n\n");
+        const mediaMarkdown = [imageMarkdown, videoMarkdown].filter(Boolean).join("\n\n");
+        if (mediaMarkdown) {
+          finalBody = finalBody + "\n\n" + mediaMarkdown;
+        }
+      }
+    }
+    const markdown = `---
+title: "${title.replace(/"/g, '\\"')}"
+date: ${date}
+lang: ${lang}
+original_link: "${post.link || ""}"
+${isOriginal ? "" : `translated_from: "${translate?.sourceLang || "ru"}"
+`}---
+
+${finalBody}
+`;
+    fs2.writeFileSync(path2.join(postDir, "index.md"), markdown);
+    createdPosts.push(`${lang}/${slug}`);
+    onProgress?.(`  Created: ${lang}/${slug}`);
+  }
+  return createdPosts;
+}
+async function exportAndTranslate(options) {
+  const {
+    apiId,
+    apiHash,
+    session,
+    channel,
+    outputDir,
+    limit,
+    since,
+    downloadMedia = true,
+    onProgress,
+    onSession,
+    processedLog = {},
+    onProcessedLog
+  } = options;
+  const exportDir = path2.join(outputDir, "..", ".telegram-export");
+  fs2.mkdirSync(exportDir, { recursive: true });
+  fs2.mkdirSync(outputDir, { recursive: true });
+  onProgress?.("Step 1: Exporting from Telegram...");
+  const exportResult = await exportTelegramChannel({
+    apiId,
+    apiHash,
+    session,
+    target: channel,
+    outputDir: exportDir,
+    limit,
+    since,
+    downloadMedia,
+    onProgress: (current, total, msg) => {
+      onProgress?.(`  ${msg} (${current}/${total})`);
+    },
+    onSession
+  });
+  onProgress?.(`  Exported ${exportResult.posts.length} posts from "${exportResult.channelMeta.title}"`);
+  onProgress?.("\nStep 2: Processing posts...");
+  let processed = 0;
+  let skipped = 0;
+  for (const post of exportResult.posts) {
+    const postId = `${post.channelUsername}-${post.msgId}`;
+    if (processedLog[postId]) {
+      skipped++;
+      continue;
+    }
+    if (!post.content && !post.hasMedia) {
+      skipped++;
+      continue;
+    }
+    onProgress?.(`
+Processing: ${postId}`);
+    try {
+      const created = await processPost(post, options, exportDir);
+      if (created.length > 0) {
+        processed++;
+        processedLog[postId] = {
+          posts: created,
+          date: post.date.toISOString(),
+          processedAt: (/* @__PURE__ */ new Date()).toISOString()
+        };
+        onProcessedLog?.(processedLog);
+      } else {
+        skipped++;
+      }
+    } catch (error) {
+      onProgress?.(`  Error: ${error.message}`);
+      skipped++;
+    }
+    await new Promise((r) => setTimeout(r, 200));
+  }
+  return {
+    exported: exportResult.posts.length,
+    processed,
+    skipped,
+    channelMeta: exportResult.channelMeta,
+    session: exportResult.session
+  };
+}
 // Annotate the CommonJS export names for ESM import in node:
 0 && (module.exports = {
   categorizePost,
   cleanContent,
   configureAnalytics,
   deduplicatePosts,
+  exportAndTranslate,
   exportTelegramChannel,
   extractAttachments,
   extractExcerpt,

package/dist/index.mjs

@@ -530,11 +530,252 @@ async function resumeExport(options) {
   }
   return exportTelegramChannel(options);
 }
+
+// src/exporter.ts
+import * as fs2 from "fs";
+import * as path2 from "path";
+var translitMap2 = {
+  "\u0430": "a",
+  "\u0431": "b",
+  "\u0432": "v",
+  "\u0433": "g",
+  "\u0434": "d",
+  "\u0435": "e",
+  "\u0451": "yo",
+  "\u0436": "zh",
+  "\u0437": "z",
+  "\u0438": "i",
+  "\u0439": "y",
+  "\u043A": "k",
+  "\u043B": "l",
+  "\u043C": "m",
+  "\u043D": "n",
+  "\u043E": "o",
+  "\u043F": "p",
+  "\u0440": "r",
+  "\u0441": "s",
+  "\u0442": "t",
+  "\u0443": "u",
+  "\u0444": "f",
+  "\u0445": "h",
+  "\u0446": "ts",
+  "\u0447": "ch",
+  "\u0448": "sh",
+  "\u0449": "sch",
+  "\u044A": "",
+  "\u044B": "y",
+  "\u044C": "",
+  "\u044D": "e",
+  "\u044E": "yu",
+  "\u044F": "ya"
+};
+function transliterate(text) {
+  return text.toLowerCase().split("").map((char) => translitMap2[char] || char).join("");
+}
+function generateSlug2(text, lang) {
+  let processed = text;
+  if (/[а-яё]/i.test(text)) {
+    processed = transliterate(text);
+  }
+  return processed.toLowerCase().replace(/[^a-z0-9\s-]/g, "").replace(/\s+/g, "-").replace(/-+/g, "-").replace(/^-|-$/g, "").substring(0, 60) || "untitled";
+}
+function cleanContent2(text) {
+  return text.replace(/##\s*Attachments\n(?:- [^\n]+\n?)*/g, "").replace(/- media\/\d+\/[^\n]+/g, "").replace(/#\w+@\w+/g, "").trim();
+}
+function extractTitleAndBody(content) {
+  const lines = content.split("\n").filter((l) => l.trim());
+  const title = lines[0]?.replace(/[#@[\]*]/g, "").trim() || "Untitled";
+  const body = lines.slice(1).join("\n").trim();
+  return { title, body };
+}
+async function processPost(post, options, exportDir) {
+  const { translate, outputDir, mediaDir, onProgress } = options;
+  const createdPosts = [];
+  const cleanedContent = cleanContent2(post.content || "");
+  const { title: originalTitle, body: originalBody } = extractTitleAndBody(cleanedContent);
+  const date = post.date.toISOString().split("T")[0];
+  const languages = [];
+  if (translate && translate.targetLangs.length > 0) {
+    const sourceLang = translate.sourceLang || "ru";
+    if (translate.keepOriginal) {
+      languages.push({
+        lang: sourceLang,
+        title: originalTitle,
+        body: originalBody,
+        isOriginal: true
+      });
+    }
+    for (const targetLang of translate.targetLangs) {
+      if (targetLang === sourceLang) continue;
+      onProgress?.(`  Translating to ${targetLang}...`);
+      const translateOpts = {
+        apiKey: translate.apiKey,
+        apiUrl: translate.apiUrl,
+        model: translate.model,
+        sourceLang,
+        targetLang
+      };
+      try {
+        const translatedTitle = await translateTitle(originalTitle, translateOpts);
+        const translatedBody = originalBody ? await translateContent(originalBody, translateOpts) : "";
+        languages.push({
+          lang: targetLang,
+          title: translatedTitle,
+          body: translatedBody,
+          isOriginal: false
+        });
+        onProgress?.(`  \u2192 ${targetLang}: "${translatedTitle.substring(0, 40)}..."`);
+      } catch (error) {
+        onProgress?.(`  Error translating to ${targetLang}: ${error.message}`);
+      }
+      await new Promise((r) => setTimeout(r, 300));
+    }
+  } else {
+    languages.push({
+      lang: "original",
+      title: originalTitle,
+      body: originalBody,
+      isOriginal: true
+    });
+  }
+  for (const { lang, title, body, isOriginal } of languages) {
+    const slug = generateSlug2(title, lang);
+    let postDir;
+    if (languages.length > 1 || translate && translate.targetLangs.length > 0) {
+      postDir = path2.join(outputDir, lang, slug);
+    } else {
+      postDir = path2.join(outputDir, slug);
+    }
+    if (fs2.existsSync(postDir)) {
+      onProgress?.(`  Skipping existing: ${lang}/${slug}`);
+      continue;
+    }
+    fs2.mkdirSync(postDir, { recursive: true });
+    let finalBody = body;
+    const mediaFiles = [];
+    if (post.mediaFiles && post.mediaFiles.length > 0) {
+      const paddedId = String(post.msgId).padStart(6, "0");
+      const targetMediaDir = mediaDir ? path2.join(mediaDir, paddedId) : path2.join(outputDir, "..", "media", paddedId);
+      fs2.mkdirSync(targetMediaDir, { recursive: true });
+      for (let i = 0; i < post.mediaFiles.length; i++) {
+        const sourcePath = path2.join(exportDir, post.mediaFiles[i]);
+        const ext = path2.extname(sourcePath).toLowerCase() || ".jpg";
+        const newName = `image${i + 1}${ext}`;
+        const targetPath = path2.join(targetMediaDir, newName);
+        if (fs2.existsSync(sourcePath) && !fs2.existsSync(targetPath)) {
+          fs2.copyFileSync(sourcePath, targetPath);
+          mediaFiles.push(newName);
+        }
+      }
+      if (mediaFiles.length > 0) {
+        const images = mediaFiles.filter((f) => !f.match(/\.(mp4|mov|webm|m4v)$/i));
+        const videos = mediaFiles.filter((f) => f.match(/\.(mp4|mov|webm|m4v)$/i));
+        const imageMarkdown = images.map((f) => `![](/media/${paddedId}/${f})`).join("\n\n");
+        const videoMarkdown = videos.map((f) => `<video src="/media/${paddedId}/${f}" controls></video>`).join("\n\n");
+        const mediaMarkdown = [imageMarkdown, videoMarkdown].filter(Boolean).join("\n\n");
+        if (mediaMarkdown) {
+          finalBody = finalBody + "\n\n" + mediaMarkdown;
+        }
+      }
+    }
+    const markdown = `---
+title: "${title.replace(/"/g, '\\"')}"
+date: ${date}
+lang: ${lang}
+original_link: "${post.link || ""}"
+${isOriginal ? "" : `translated_from: "${translate?.sourceLang || "ru"}"
+`}---
+
+${finalBody}
+`;
+    fs2.writeFileSync(path2.join(postDir, "index.md"), markdown);
+    createdPosts.push(`${lang}/${slug}`);
+    onProgress?.(`  Created: ${lang}/${slug}`);
+  }
+  return createdPosts;
+}
+async function exportAndTranslate(options) {
+  const {
+    apiId,
+    apiHash,
+    session,
+    channel,
+    outputDir,
+    limit,
+    since,
+    downloadMedia = true,
+    onProgress,
+    onSession,
+    processedLog = {},
+    onProcessedLog
+  } = options;
+  const exportDir = path2.join(outputDir, "..", ".telegram-export");
+  fs2.mkdirSync(exportDir, { recursive: true });
+  fs2.mkdirSync(outputDir, { recursive: true });
+  onProgress?.("Step 1: Exporting from Telegram...");
+  const exportResult = await exportTelegramChannel({
+    apiId,
+    apiHash,
+    session,
+    target: channel,
+    outputDir: exportDir,
+    limit,
+    since,
+    downloadMedia,
+    onProgress: (current, total, msg) => {
+      onProgress?.(`  ${msg} (${current}/${total})`);
+    },
+    onSession
+  });
+  onProgress?.(`  Exported ${exportResult.posts.length} posts from "${exportResult.channelMeta.title}"`);
+  onProgress?.("\nStep 2: Processing posts...");
+  let processed = 0;
+  let skipped = 0;
+  for (const post of exportResult.posts) {
+    const postId = `${post.channelUsername}-${post.msgId}`;
+    if (processedLog[postId]) {
+      skipped++;
+      continue;
+    }
+    if (!post.content && !post.hasMedia) {
+      skipped++;
+      continue;
+    }
+    onProgress?.(`
+Processing: ${postId}`);
+    try {
+      const created = await processPost(post, options, exportDir);
+      if (created.length > 0) {
+        processed++;
+        processedLog[postId] = {
+          posts: created,
+          date: post.date.toISOString(),
+          processedAt: (/* @__PURE__ */ new Date()).toISOString()
+        };
+        onProcessedLog?.(processedLog);
+      } else {
+        skipped++;
+      }
+    } catch (error) {
+      onProgress?.(`  Error: ${error.message}`);
+      skipped++;
+    }
+    await new Promise((r) => setTimeout(r, 200));
+  }
+  return {
+    exported: exportResult.posts.length,
+    processed,
+    skipped,
+    channelMeta: exportResult.channelMeta,
+    session: exportResult.session
+  };
+}
 export {
   categorizePost,
   cleanContent,
   configureAnalytics,
   deduplicatePosts,
+  exportAndTranslate,
   exportTelegramChannel,
   extractAttachments,
   extractExcerpt,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "koztv-blog-tools",
-  "version": "1.1.1",
+  "version": "1.2.1",
   "description": "Shared utilities for Telegram-based blog sites",
   "main": "dist/index.js",
   "module": "dist/index.mjs",