teledzik 1.0.2 → 1.0.3

package/lib/core/helpers/rich-sanitizer.js

@@ -1,30 +1,100 @@
 "use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.sanitizeRichHtml = void 0;
 /**
- * Converts Rich HTML strings containing tags like <h1>, <h2>, <p>, <hr>
+ * Converts Rich HTML strings (containing unsupported Telegram HTML tags like
+ * <table>, <ul>, <ol>, <li>, <details>, <h1>-<h6>, <aside>, <p>, <hr>, etc.)
  * into standard Telegram HTML compatible with parse_mode: 'HTML' (editMessageText/sendMessage).
  */
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.sanitizeRichHtml = void 0;
+function convertTableToText(tableHtml) {
+    const rows = [];
+    const rowMatches = tableHtml.match(/<tr[^>]*>([\s\S]*?)<\/tr>/gi) || [];
+    for (const row of rowMatches) {
+        const cells = [];
+        const cellMatches = row.match(/<(th|td)[^>]*>([\s\S]*?)<\/\1>/gi) || [];
+        for (const cell of cellMatches) {
+            const content = cell.replace(/<(th|td)[^>]*>([\s\S]*?)<\/\1>/i, '$2').trim();
+            cells.push(content);
+        }
+        if (cells.length > 0)
+            rows.push(cells);
+    }
+    if (rows.length === 0)
+        return '';
+    // Calculate column widths
+    const colWidths = [];
+    rows.forEach((r) => {
+        r.forEach((c, idx) => {
+            const textLen = c.replace(/<[^>]+>/g, '').length;
+            colWidths[idx] = Math.max(colWidths[idx] || 0, textLen);
+        });
+    });
+    // Format table rows
+    const formattedRows = rows.map((r, rowIdx) => {
+        const line = r
+            .map((c, colIdx) => {
+            const textLen = c.replace(/<[^>]+>/g, '').length;
+            const pad = ' '.repeat(Math.max(0, (colWidths[colIdx] || 0) - textLen));
+            return `${c}${pad}`;
+        })
+            .join(' │ ');
+        if (rowIdx === 0 && rows.length > 1) {
+            const sep = colWidths.map((w) => '─'.repeat(w)).join('─┼─');
+            return `${line}\n${sep}`;
+        }
+        return line;
+    });
+    return `<pre>\n${formattedRows.join('\n')}\n</pre>\n\n`;
+}
 function sanitizeRichHtml(html) {
     if (!html)
         return html;
     let result = html;
-    // Convert H1: <h1>Text</h1> -> <b>TEXT</b>
-    result = result.replace(/<h1[^>]*>(.*?)<\/h1>/gis, (_, content) => `<b>${content.trim().toUpperCase()}</b>\n\n`);
-    // Convert H2..H6: <h2>Text</h2> -> <b>Text</b>
-    result = result.replace(/<h[2-6][^>]*>(.*?)<\/h[2-6]>/gis, (_, content) => `<b>${content.trim()}</b>\n\n`);
-    // Convert Paragraphs: <p>Text</p> -> Text\n\n
-    result = result.replace(/<p[^>]*>(.*?)<\/p>/gis, (_, content) => `${content.trim()}\n\n`);
-    // Convert Footer: <footer>Text</footer> -> <i>Text</i>
-    result = result.replace(/<footer[^>]*>(.*?)<\/footer>/gis, (_, content) => `<i>${content.trim()}</i>\n`);
-    // Convert Horizontal Rules: <hr/> or <hr> -> ──────────
+    // 1. Convert Tables: <table>...</table> -> <pre>formatted table</pre>
+    result = result.replace(/<table[^>]*>([\s\S]*?)<\/table>/gi, (_, content) => convertTableToText(content));
+    // 2. Convert Details / Expandable: <details><summary>Title</summary>Content</details>
+    result = result.replace(/<details[^>]*>\s*<summary[^>]*>(.*?)<\/summary>([\s\S]*?)<\/details>/gi, (_, title, body) => {
+        return `<b>▶ ${title.trim()}</b>\n<blockquote>${body.trim()}</blockquote>\n\n`;
+    });
+    // 3. Convert Headings
+    result = result.replace(/<h1[^>]*>([\s\S]*?)<\/h1>/gi, (_, c) => `<b>${c.trim().toUpperCase()}</b>\n\n`);
+    result = result.replace(/<h[2-6][^>]*>([\s\S]*?)<\/h[2-6]>/gi, (_, c) => `<b>${c.trim()}</b>\n\n`);
+    // 4. Convert Task Lists & Unordered/Ordered Lists
+    result = result.replace(/<ul[^>]*>([\s\S]*?)<\/ul>/gi, (_, listContent) => {
+        return (listContent
+            .replace(/<li[^>]*>\s*<input[^>]*type="checkbox"[^>]*checked[^>]*>\s*([\s\S]*?)<\/li>/gi, '☑ $1\n')
+            .replace(/<li[^>]*>\s*<input[^>]*type="checkbox"[^>]*>\s*([\s\S]*?)<\/li>/gi, '☐ $1\n')
+            .replace(/<li[^>]*>([\s\S]*?)<\/li>/gi, '• $1\n')
+            .trim() + '\n\n');
+    });
+    result = result.replace(/<ol[^>]*>([\s\S]*?)<\/ol>/gi, (_, listContent) => {
+        let index = 1;
+        return (listContent
+            .replace(/<li[^>]*>([\s\S]*?)<\/li>/gi, (_match, item) => `${index++}. ${item.trim()}\n`)
+            .trim() + '\n\n');
+    });
+    // 5. Convert Pull Quotes / Asides
+    result = result.replace(/<aside[^>]*>([\s\S]*?)(?:<cite[^>]*>([\s\S]*?)<\/cite>)?<\/aside>/gi, (_, quote, cite) => {
+        return `<blockquote>${quote.trim()}${cite ? `\n— <i>${cite.trim()}</i>` : ''}</blockquote>\n\n`;
+    });
+    // 6. Convert Special Bot API 10.1 tags
+    result = result.replace(/<tg-thinking[^>]*>([\s\S]*?)<\/tg-thinking>/gi, (_, c) => `<blockquote>💭 <i>${c.trim()}</i></blockquote>\n\n`);
+    result = result.replace(/<tg-math-block[^>]*>([\s\S]*?)<\/tg-math-block>/gi, (_, c) => `<pre><code>${c.trim()}</code></pre>\n\n`);
+    result = result.replace(/<tg-reference[^>]*>([\s\S]*?)<\/tg-reference>/gi, (_, c) => `<i>${c.trim()}</i>`);
+    result = result.replace(/<mark[^>]*>([\s\S]*?)<\/mark>/gi, (_, c) => `<b>[${c.trim()}]</b>`);
+    // 7. Convert Figures & Captions
+    result = result.replace(/<figure[^>]*>[\s\S]*?<figcaption[^>]*>([\s\S]*?)<\/figcaption><\/figure>/gi, (_, cap) => `\n<i>📷 ${cap.trim()}</i>\n`);
+    // 8. Convert Paragraphs & Linebreaks
+    result = result.replace(/<p[^>]*>([\s\S]*?)<\/p>/gi, (_, c) => `${c.trim()}\n\n`);
+    result = result.replace(/<footer[^>]*>([\s\S]*?)<\/footer>/gi, (_, c) => `<i>${c.trim()}</i>\n`);
+    result = result.replace(/<br\s*\/?>/gi, '\n');
     result = result.replace(/<hr\s*\/?>/gi, '──────────\n');
-    // Convert Marks: <mark>Text</mark> -> <b>[Text]</b>
-    result = result.replace(/<mark[^>]*>(.*?)<\/mark>/gis, (_, content) => `<b>[${content.trim()}]</b>`);
-    // Strip unsupported sub/sup tags but keep text
-    result = result.replace(/<sub[^>]*>(.*?)<\/sub>/gis, '$1');
-    result = result.replace(/<sup[^>]*>(.*?)<\/sup>/gis, '$1');
-    // Trim extra trailing newlines (> 2 consecutive newlines)
+    // 9. Strip unsupported media & sub/sup tags but keep text
+    result = result.replace(/<sub[^>]*>([\s\S]*?)<\/sub>/gi, '$1');
+    result = result.replace(/<sup[^>]*>([\s\S]*?)<\/sup>/gi, '$1');
+    result = result.replace(/<tg-(map|collage|slideshow)[^>]*>([\s\S]*?)<\/tg-\1>/gi, '$2');
+    result = result.replace(/<(img|video|audio|tg-map)[^>]*\/?>/gi, '');
+    // 10. Clean up extra newlines (> 2 consecutive newlines)
     result = result.replace(/\n{3,}/g, '\n\n');
     return result.trim();
 }

package/lib/telegram.js

@@ -105,6 +105,9 @@ class Telegram extends client_1.default {
      */
     sendMessage(chatId, text, extra) {
         const t = format_1.FmtString.normalise(text);
+        if (extra?.parse_mode === 'HTML' && typeof t.text === 'string') {
+            t.text = (0, rich_sanitizer_1.sanitizeRichHtml)(t.text);
+        }
         return this.callApi('sendMessage', { chat_id: chatId, ...extra, ...t });
     }
     /**

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "teledzik",
-  "version": "1.0.2",
+  "version": "1.0.3",
   "description": "Modern Telegram Bot Framework",
   "keywords": [
     "telekaf",

package/src/core/helpers/rich-sanitizer.ts

@@ -1,35 +1,118 @@
 /**
- * Converts Rich HTML strings containing tags like <h1>, <h2>, <p>, <hr>
+ * Converts Rich HTML strings (containing unsupported Telegram HTML tags like
+ * <table>, <ul>, <ol>, <li>, <details>, <h1>-<h6>, <aside>, <p>, <hr>, etc.)
  * into standard Telegram HTML compatible with parse_mode: 'HTML' (editMessageText/sendMessage).
  */
+
+function convertTableToText(tableHtml: string): string {
+  const rows: string[][] = []
+  const rowMatches = tableHtml.match(/<tr[^>]*>([\s\S]*?)<\/tr>/gi) || []
+
+  for (const row of rowMatches) {
+    const cells: string[] = []
+    const cellMatches = row.match(/<(th|td)[^>]*>([\s\S]*?)<\/\1>/gi) || []
+    for (const cell of cellMatches) {
+      const content = cell.replace(/<(th|td)[^>]*>([\s\S]*?)<\/\1>/i, '$2').trim()
+      cells.push(content)
+    }
+    if (cells.length > 0) rows.push(cells)
+  }
+
+  if (rows.length === 0) return ''
+
+  // Calculate column widths
+  const colWidths: number[] = []
+  rows.forEach((r) => {
+    r.forEach((c, idx) => {
+      const textLen = c.replace(/<[^>]+>/g, '').length
+      colWidths[idx] = Math.max(colWidths[idx] || 0, textLen)
+    })
+  })
+
+  // Format table rows
+  const formattedRows = rows.map((r, rowIdx) => {
+    const line = r
+      .map((c, colIdx) => {
+        const textLen = c.replace(/<[^>]+>/g, '').length
+        const pad = ' '.repeat(Math.max(0, (colWidths[colIdx] || 0) - textLen))
+        return `${c}${pad}`
+      })
+      .join(' │ ')
+
+    if (rowIdx === 0 && rows.length > 1) {
+      const sep = colWidths.map((w) => '─'.repeat(w)).join('─┼─')
+      return `${line}\n${sep}`
+    }
+    return line
+  })
+
+  return `<pre>\n${formattedRows.join('\n')}\n</pre>\n\n`
+}
+
 export function sanitizeRichHtml(html: string): string {
   if (!html) return html
 
   let result = html
 
-  // Convert H1: <h1>Text</h1> -> <b>TEXT</b>
-  result = result.replace(/<h1[^>]*>(.*?)<\/h1>/gis, (_, content) => `<b>${content.trim().toUpperCase()}</b>\n\n`)
+  // 1. Convert Tables: <table>...</table> -> <pre>formatted table</pre>
+  result = result.replace(/<table[^>]*>([\s\S]*?)<\/table>/gi, (_, content) => convertTableToText(content))
 
-  // Convert H2..H6: <h2>Text</h2> -> <b>Text</b>
-  result = result.replace(/<h[2-6][^>]*>(.*?)<\/h[2-6]>/gis, (_, content) => `<b>${content.trim()}</b>\n\n`)
+  // 2. Convert Details / Expandable: <details><summary>Title</summary>Content</details>
+  result = result.replace(/<details[^>]*>\s*<summary[^>]*>(.*?)<\/summary>([\s\S]*?)<\/details>/gi, (_, title, body) => {
+    return `<b>▶ ${title.trim()}</b>\n<blockquote>${body.trim()}</blockquote>\n\n`
+  })
 
-  // Convert Paragraphs: <p>Text</p> -> Text\n\n
-  result = result.replace(/<p[^>]*>(.*?)<\/p>/gis, (_, content) => `${content.trim()}\n\n`)
+  // 3. Convert Headings
+  result = result.replace(/<h1[^>]*>([\s\S]*?)<\/h1>/gi, (_, c) => `<b>${c.trim().toUpperCase()}</b>\n\n`)
+  result = result.replace(/<h[2-6][^>]*>([\s\S]*?)<\/h[2-6]>/gi, (_, c) => `<b>${c.trim()}</b>\n\n`)
 
-  // Convert Footer: <footer>Text</footer> -> <i>Text</i>
-  result = result.replace(/<footer[^>]*>(.*?)<\/footer>/gis, (_, content) => `<i>${content.trim()}</i>\n`)
+  // 4. Convert Task Lists & Unordered/Ordered Lists
+  result = result.replace(/<ul[^>]*>([\s\S]*?)<\/ul>/gi, (_, listContent) => {
+    return (
+      listContent
+        .replace(/<li[^>]*>\s*<input[^>]*type="checkbox"[^>]*checked[^>]*>\s*([\s\S]*?)<\/li>/gi, '☑ $1\n')
+        .replace(/<li[^>]*>\s*<input[^>]*type="checkbox"[^>]*>\s*([\s\S]*?)<\/li>/gi, '☐ $1\n')
+        .replace(/<li[^>]*>([\s\S]*?)<\/li>/gi, '• $1\n')
+        .trim() + '\n\n'
+    )
+  })
 
-  // Convert Horizontal Rules: <hr/> or <hr> -> ──────────
-  result = result.replace(/<hr\s*\/?>/gi, '──────────\n')
+  result = result.replace(/<ol[^>]*>([\s\S]*?)<\/ol>/gi, (_: string, listContent: string) => {
+    let index = 1
+    return (
+      listContent
+        .replace(/<li[^>]*>([\s\S]*?)<\/li>/gi, (_match: string, item: string) => `${index++}. ${item.trim()}\n`)
+        .trim() + '\n\n'
+    )
+  })
+
+  // 5. Convert Pull Quotes / Asides
+  result = result.replace(/<aside[^>]*>([\s\S]*?)(?:<cite[^>]*>([\s\S]*?)<\/cite>)?<\/aside>/gi, (_, quote, cite) => {
+    return `<blockquote>${quote.trim()}${cite ? `\n— <i>${cite.trim()}</i>` : ''}</blockquote>\n\n`
+  })
 
-  // Convert Marks: <mark>Text</mark> -> <b>[Text]</b>
-  result = result.replace(/<mark[^>]*>(.*?)<\/mark>/gis, (_, content) => `<b>[${content.trim()}]</b>`)
+  // 6. Convert Special Bot API 10.1 tags
+  result = result.replace(/<tg-thinking[^>]*>([\s\S]*?)<\/tg-thinking>/gi, (_, c) => `<blockquote>💭 <i>${c.trim()}</i></blockquote>\n\n`)
+  result = result.replace(/<tg-math-block[^>]*>([\s\S]*?)<\/tg-math-block>/gi, (_, c) => `<pre><code>${c.trim()}</code></pre>\n\n`)
+  result = result.replace(/<tg-reference[^>]*>([\s\S]*?)<\/tg-reference>/gi, (_, c) => `<i>${c.trim()}</i>`)
+  result = result.replace(/<mark[^>]*>([\s\S]*?)<\/mark>/gi, (_, c) => `<b>[${c.trim()}]</b>`)
+
+  // 7. Convert Figures & Captions
+  result = result.replace(/<figure[^>]*>[\s\S]*?<figcaption[^>]*>([\s\S]*?)<\/figcaption><\/figure>/gi, (_, cap) => `\n<i>📷 ${cap.trim()}</i>\n`)
+
+  // 8. Convert Paragraphs & Linebreaks
+  result = result.replace(/<p[^>]*>([\s\S]*?)<\/p>/gi, (_, c) => `${c.trim()}\n\n`)
+  result = result.replace(/<footer[^>]*>([\s\S]*?)<\/footer>/gi, (_, c) => `<i>${c.trim()}</i>\n`)
+  result = result.replace(/<br\s*\/?>/gi, '\n')
+  result = result.replace(/<hr\s*\/?>/gi, '──────────\n')
 
-  // Strip unsupported sub/sup tags but keep text
-  result = result.replace(/<sub[^>]*>(.*?)<\/sub>/gis, '$1')
-  result = result.replace(/<sup[^>]*>(.*?)<\/sup>/gis, '$1')
+  // 9. Strip unsupported media & sub/sup tags but keep text
+  result = result.replace(/<sub[^>]*>([\s\S]*?)<\/sub>/gi, '$1')
+  result = result.replace(/<sup[^>]*>([\s\S]*?)<\/sup>/gi, '$1')
+  result = result.replace(/<tg-(map|collage|slideshow)[^>]*>([\s\S]*?)<\/tg-\1>/gi, '$2')
+  result = result.replace(/<(img|video|audio|tg-map)[^>]*\/?>/gi, '')
 
-  // Trim extra trailing newlines (> 2 consecutive newlines)
+  // 10. Clean up extra newlines (> 2 consecutive newlines)
   result = result.replace(/\n{3,}/g, '\n\n')
 
   return result.trim()

package/src/telegram.ts

@@ -138,6 +138,9 @@ export class Telegram extends ApiClient {
     extra?: tt.ExtraReplyMessage
   ) {
     const t = FmtString.normalise(text)
+    if (extra?.parse_mode === 'HTML' && typeof t.text === 'string') {
+      t.text = sanitizeRichHtml(t.text)
+    }
     return this.callApi('sendMessage', { chat_id: chatId, ...extra, ...t })
   }
 

package/typings/core/helpers/rich-sanitizer.d.ts

@@ -1,5 +1,6 @@
 /**
- * Converts Rich HTML strings containing tags like <h1>, <h2>, <p>, <hr>
+ * Converts Rich HTML strings (containing unsupported Telegram HTML tags like
+ * <table>, <ul>, <ol>, <li>, <details>, <h1>-<h6>, <aside>, <p>, <hr>, etc.)
  * into standard Telegram HTML compatible with parse_mode: 'HTML' (editMessageText/sendMessage).
  */
 export declare function sanitizeRichHtml(html: string): string;