Initial commit

importers/telegram_html.py

@@ -0,0 +1,66 @@
+from __future__ import annotations
+
+import re
+from pathlib import Path
+from typing import Dict, Iterator, List, Optional
+
+from bs4 import BeautifulSoup
+
+RESUME_EXTS = {".pdf", ".docx", ".doc", ".txt", ".html", ".htm"}
+
+def find_messages_html(root: Path) -> List[Path]:
+    return [p for p in root.rglob("messages*.html") if p.is_file()]
+
+def iter_artifacts(messages_html: Path) -> Iterator[Dict]:
+    html = messages_html.read_text(encoding="utf-8", errors="ignore")
+    soup = BeautifulSoup(html, "lxml")
+
+    chat_title = None
+    h = soup.find(class_=re.compile(r"page_header", re.I))
+    if h:
+        chat_title = h.get_text(" ", strip=True)
+    chat_title = chat_title or messages_html.parent.name
+
+    for msg in soup.select(".message.default.clearfix, .message"):
+        message_id = msg.get("id") or None
+        date_div = msg.select_one(".date")
+        msg_date = date_div.get("title") if date_div else None
+
+        text_div = msg.select_one(".text")
+        msg_text = text_div.get_text("\n", strip=True) if text_div else ""
+
+        file_path = None
+        original_name = None
+        for a in msg.find_all("a", href=True):
+            href = a["href"]
+            p = (messages_html.parent / href).resolve()
+            if p.exists() and p.suffix.lower() in RESUME_EXTS:
+                file_path = str(p)
+                original_name = p.name
+                break
+
+        if file_path:
+            yield {
+                "origin_type": "telegram_html",
+                "export_path": str(messages_html.parent),
+                "chat_title": chat_title,
+                "message_id": str(message_id) if message_id else None,
+                "message_date": msg_date,
+                "message_text": msg_text or "",
+                "file_path": file_path,
+                "original_name": original_name,
+                "extra": {"html_path": str(messages_html)},
+            }
+        else:
+            if msg_text and len(msg_text.strip()) >= 500:
+                yield {
+                    "origin_type": "message_text",
+                    "export_path": str(messages_html.parent),
+                    "chat_title": chat_title,
+                    "message_id": str(message_id) if message_id else None,
+                    "message_date": msg_date,
+                    "message_text": msg_text,
+                    "file_path": None,
+                    "original_name": None,
+                    "extra": {"html_path": str(messages_html)},
+                }