From 1b64991041ffec8c89e75c2a1dfd9ab3f8dc81f6 Mon Sep 17 00:00:00 2001
From: pacnpal <183241239+pacnpal@users.noreply.github.com>
Date: Fri, 15 Nov 2024 02:03:23 +0000
Subject: [PATCH] fix: Change URL extraction in processor to use regex

- Replaced extractor.suitable() with direct regex pattern matching
- Added check for _VALID_URL attribute
- Improved error handling in URL extraction
- Maintains compatibility with video_downloader.py changes
---
 videoarchiver/processor.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/videoarchiver/processor.py b/videoarchiver/processor.py
index 8198d20..c008bf0 100644
--- a/videoarchiver/processor.py
+++ b/videoarchiver/processor.py
@@ -293,11 +293,11 @@ class VideoProcessor:
                 for word in words:
                     # Try each extractor
                     for ie in ydl._ies:
-                        # Use suitable as a classmethod and check the result
-                        result = ie.suitable(word)
-                        if result and not isinstance(result, str):
-                            urls.append(word)
-                            break  # Stop once we find a matching extractor
+                        if hasattr(ie, '_VALID_URL') and ie._VALID_URL:
+                            # Use regex pattern matching instead of suitable()
+                            if re.match(ie._VALID_URL, word):
+                                urls.append(word)
+                                break  # Stop once we find a matching pattern
         except Exception as e:
             logger.error(f"URL extraction error: {str(e)}")
         return list(set(urls))  # Remove duplicates