commit

src/core/api_client.py

@@ -41,9 +41,14 @@ def fetch_posts_paginated(api_url_base, headers, offset, logger, cancellation_ev
         try:
             response = requests.get(paginated_url, headers=headers, timeout=(15, 60), cookies=cookies_dict)
             response.raise_for_status()
+            response.encoding = 'utf-8'  
             return response.json()
 
         except requests.exceptions.RequestException as e:
+            if e.response is not None and e.response.status_code == 400:
+                logger(f"   ✅ Reached end of posts (API returned 400 Bad Request for offset {offset}).")
+                return [] 
+
             logger(f"   ⚠️ Retryable network error on page fetch (Attempt {attempt + 1}): {e}")
             if attempt < max_retries - 1:
                 delay = retry_delay * (2 ** attempt)
@@ -81,8 +86,11 @@ def fetch_single_post_data(api_domain, service, user_id, post_id, headers, logge
                 response_body += chunk
             
             full_post_data = json.loads(response_body)
+
             if isinstance(full_post_data, list) and full_post_data:
                 return full_post_data[0] 
+            if isinstance(full_post_data, dict) and 'post' in full_post_data:
+                return full_post_data['post'] 
             return full_post_data 
             
     except Exception as e:
@@ -101,6 +109,7 @@ def fetch_post_comments(api_domain, service, user_id, post_id, headers, logger,
     try:
         response = requests.get(comments_api_url, headers=headers, timeout=(10, 30), cookies=cookies_dict)
         response.raise_for_status()
+        response.encoding = 'utf-8'          
         return response.json()
     except requests.exceptions.RequestException as e:
         raise RuntimeError(f"Error fetching comments for post {post_id}: {e}")
@@ -141,12 +150,9 @@ def download_from_api(
     parsed_input_url_for_domain = urlparse(api_url_input)
     api_domain = parsed_input_url_for_domain.netloc
     
-    # --- START: MODIFIED LOGIC ---
-    # This list is updated to include the new .cr and .st mirrors for validation.
     if not any(d in api_domain.lower() for d in ['kemono.su', 'kemono.party', 'kemono.cr', 'coomer.su', 'coomer.party', 'coomer.st']):
         logger(f"⚠️ Unrecognized domain '{api_domain}' from input URL. Defaulting to kemono.su for API calls.")
         api_domain = "kemono.su"
-    # --- END: MODIFIED LOGIC ---
         
     cookies_for_api = None
     if use_cookie and app_base_dir:
@@ -160,6 +166,7 @@ def download_from_api(
         try:
             direct_response = requests.get(direct_post_api_url, headers=headers, timeout=(10, 30), cookies=cookies_for_api)
             direct_response.raise_for_status()
+            direct_response.encoding = 'utf-8' 
             direct_post_data = direct_response.json()
             if isinstance(direct_post_data, list) and direct_post_data:
                 direct_post_data = direct_post_data[0]
@@ -185,7 +192,7 @@ def download_from_api(
 
     is_manga_mode_fetch_all_and_sort_oldest_first = manga_mode and (manga_filename_style_for_sort_check != STYLE_DATE_POST_TITLE) and not target_post_id
     should_fetch_all = fetch_all_first or is_manga_mode_fetch_all_and_sort_oldest_first  
-    api_base_url = f"https://{api_domain}/api/v1/{service}/user/{user_id}"
+    api_base_url = f"https://{api_domain}/api/v1/{service}/user/{user_id}/posts"
     page_size = 50
     if is_manga_mode_fetch_all_and_sort_oldest_first:
         logger(f"   Manga Mode (Style: {manga_filename_style_for_sort_check if manga_filename_style_for_sort_check else 'Default'} - Oldest First Sort Active): Fetching all posts to sort by date...")

src/core/workers.py

@@ -37,7 +37,7 @@ try:
 except ImportError:
     Document = None
 from PyQt5 .QtCore import Qt ,QThread ,pyqtSignal ,QMutex ,QMutexLocker ,QObject ,QTimer ,QSettings ,QStandardPaths ,QCoreApplication ,QUrl ,QSize ,QProcess 
-from .api_client import download_from_api, fetch_post_comments
+from .api_client import download_from_api, fetch_post_comments, fetch_single_post_data
 from ..services.multipart_downloader import download_file_in_parts, MULTIPART_DOWNLOADER_AVAILABLE
 from ..services.drive_downloader import (
     download_mega_file, download_gdrive_file, download_dropbox_file
@@ -124,7 +124,8 @@ class PostProcessorWorker:
                  processed_post_ids=None,
                  multipart_scope='both', 
                  multipart_parts_count=4, 
-                 multipart_min_size_mb=100 
+                 multipart_min_size_mb=100,
+                 skip_file_size_mb=None 
                  ):
         self.post = post_data
         self.download_root = download_root
@@ -189,6 +190,7 @@ class PostProcessorWorker:
         self.multipart_scope = multipart_scope 
         self.multipart_parts_count = multipart_parts_count 
         self.multipart_min_size_mb = multipart_min_size_mb 
+        self.skip_file_size_mb = skip_file_size_mb
         if self.compress_images and Image is None:
             self.logger("⚠️ Image compression disabled: Pillow library not found.")
             self.compress_images = False
@@ -277,6 +279,24 @@ class PostProcessorWorker:
         if self.use_cookie:
             cookies_to_use_for_file = prepare_cookies_for_request(self.use_cookie, self.cookie_text, self.selected_cookie_file, self.app_base_dir, self.logger)
         
+        if self.skip_file_size_mb is not None:
+                api_original_filename_for_size_check = file_info.get('_original_name_for_log', file_info.get('name'))
+                try:
+                        # Use a stream=True HEAD request to get headers without downloading the body
+                        with requests.head(file_url, headers=file_download_headers, timeout=15, cookies=cookies_to_use_for_file, allow_redirects=True) as head_response:
+                                head_response.raise_for_status()
+                                content_length = head_response.headers.get('Content-Length')
+                                if content_length:
+                                        file_size_bytes = int(content_length)
+                                        file_size_mb = file_size_bytes / (1024 * 1024)
+                                        if file_size_mb < self.skip_file_size_mb:
+                                                self.logger(f"   -> Skip File (Size): '{api_original_filename_for_size_check}' is {file_size_mb:.2f} MB, which is smaller than the {self.skip_file_size_mb} MB limit.")
+                                                return 0, 1, api_original_filename_for_size_check, False, FILE_DOWNLOAD_STATUS_SKIPPED, None
+                                else:
+                                        self.logger(f"   ⚠️ Could not determine file size for '{api_original_filename_for_size_check}' to check against size limit. Proceeding with download.")
+                except requests.RequestException as e:
+                        self.logger(f"   ⚠️ Could not fetch file headers to check size for '{api_original_filename_for_size_check}': {e}. Proceeding with download.")
+                
         api_original_filename = file_info.get('_original_name_for_log', file_info.get('name'))
         filename_to_save_in_main_path = ""
         if forced_filename_override:
@@ -488,19 +508,18 @@ class PostProcessorWorker:
                 except requests.RequestException as e:
                     self.logger(f"   ⚠️ Could not verify size of existing file '{filename_to_save_in_main_path}': {e}. Proceeding with download.")
         
+        max_retries = 3
         retry_delay = 5
         downloaded_size_bytes = 0
         calculated_file_hash = None
         downloaded_part_file_path = None
-        total_size_bytes = 0
         download_successful_flag = False
         last_exception_for_retry_later = None
         is_permanent_error = False
         data_to_write_io = None
 
-        response_for_this_attempt = None
         for attempt_num_single_stream in range(max_retries + 1):
-            response_for_this_attempt = None
+            response = None
             if self._check_pause(f"File download attempt for '{api_original_filename}'"): break
             if self.check_cancel() or (skip_event and skip_event.is_set()): break
             try:
@@ -519,12 +538,24 @@ class PostProcessorWorker:
                     new_url = self._find_valid_subdomain(current_url_to_try)
                     if new_url != current_url_to_try:
                         self.logger(f"   Retrying with new URL: {new_url}")
-                        file_url = new_url # Update the main file_url for subsequent retries
+                        file_url = new_url
+                        response.close() # Close the old response
                         response = requests.get(new_url, headers=file_download_headers, timeout=(30, 300), stream=True, cookies=cookies_to_use_for_file)
 
-
                 response.raise_for_status()
+                
+                # --- REVISED AND MOVED SIZE CHECK LOGIC ---
                 total_size_bytes = int(response.headers.get('Content-Length', 0))
+
+                if self.skip_file_size_mb is not None:
+                    if total_size_bytes > 0:
+                        file_size_mb = total_size_bytes / (1024 * 1024)
+                        if file_size_mb < self.skip_file_size_mb:
+                            self.logger(f"   -> Skip File (Size): '{api_original_filename}' is {file_size_mb:.2f} MB, which is smaller than the {self.skip_file_size_mb} MB limit.")
+                            return 0, 1, api_original_filename, False, FILE_DOWNLOAD_STATUS_SKIPPED, None
+                    # If Content-Length is missing, we can't check, so we no longer log a warning here and just proceed.
+                # --- END OF REVISED LOGIC ---
+
                 num_parts_for_file = min(self.multipart_parts_count, MAX_PARTS_FOR_MULTIPART_DOWNLOAD)
            
                 file_is_eligible_by_scope = False
@@ -548,9 +579,7 @@ class PostProcessorWorker:
                 if self._check_pause(f"Multipart decision for '{api_original_filename}'"): break
 
                 if attempt_multipart:
-                    if response_for_this_attempt:
+                    response.close() # Close the initial connection before starting multipart
-                        response_for_this_attempt.close()
-                        response_for_this_attempt = None
                     mp_save_path_for_unique_part_stem_arg = os.path.join(target_folder_path, f"{unique_part_file_stem_on_disk}{temp_file_ext_for_unique_part}")
                     mp_success, mp_bytes, mp_hash, mp_file_handle = download_file_in_parts(
                         file_url, mp_save_path_for_unique_part_stem_arg, total_size_bytes, num_parts_for_file, file_download_headers, api_original_filename,
@@ -576,7 +605,6 @@ class PostProcessorWorker:
                     current_attempt_downloaded_bytes = 0
                     md5_hasher = hashlib.md5()
                     last_progress_time = time.time()
-                    single_stream_exception = None
                     try:
                         with open(current_single_stream_part_path, 'wb') as f_part:
                             for chunk in response.iter_content(chunk_size=1 * 1024 * 1024):
@@ -643,8 +671,8 @@ class PostProcessorWorker:
                 is_permanent_error = True                
                 break
             finally:
-                if response_for_this_attempt:
+                if response:
-                    response_for_this_attempt.close()
+                    response.close()
                 self._emit_signal('file_download_status', False)
 
         final_total_for_progress = total_size_bytes if download_successful_flag and total_size_bytes > 0 else downloaded_size_bytes
@@ -848,6 +876,37 @@ class PostProcessorWorker:
             post_data = self.post  # Reference to the post object
             log_prefix = "Post"
 
+        # --- FIX: FETCH FULL POST DATA IF CONTENT IS MISSING BUT NEEDED ---
+        content_is_needed = (
+            self.show_external_links or
+            self.extract_links_only or
+            self.scan_content_for_images or
+            (self.filter_mode == 'text_only' and self.text_only_scope == 'content')
+        )
+
+        if content_is_needed and self.post.get('content') is None and self.service != 'discord':
+            self.logger(f"   Post {post_id} is missing 'content' field, fetching full data...")
+            parsed_url = urlparse(self.api_url_input)
+            api_domain = parsed_url.netloc
+            headers = {'User-Agent': 'Mozilla/5.0'}
+            cookies = prepare_cookies_for_request(self.use_cookie, self.cookie_text, self.selected_cookie_file, self.app_base_dir, self.logger, target_domain=api_domain)
+
+            full_post_data = fetch_single_post_data(api_domain, self.service, self.user_id, post_id, headers, self.logger, cookies_dict=cookies)
+            
+            if full_post_data:
+                self.logger("   ✅ Full post data fetched successfully.")
+                # Update the worker's post object with the complete data
+                self.post = full_post_data
+                # Re-initialize local variables from the new, complete post data
+                post_title = self.post.get('title', '') or 'untitled_post'
+                post_main_file_info = self.post.get('file')
+                post_attachments = self.post.get('attachments', [])
+                post_content_html = self.post.get('content', '')
+                post_data = self.post
+            else:
+                self.logger(f"   ⚠️ Failed to fetch full content for post {post_id}. Content-dependent features may not work for this post.")
+        # --- END FIX ---
+
         # 2. SHARED PROCESSING LOGIC: The rest of the function now uses the consistent variables from above.
         result_tuple = (0, 0, [], [], [], None, None)
         total_downloaded_this_post = 0
@@ -1258,7 +1317,6 @@ class PostProcessorWorker:
                     parsed_url = urlparse(self.api_url_input)
                     api_domain = parsed_url.netloc
                     cookies = prepare_cookies_for_request(self.use_cookie, self.cookie_text, self.selected_cookie_file, self.app_base_dir, self.logger, target_domain=api_domain)
-                    from .api_client import fetch_single_post_data
                     full_data = fetch_single_post_data(api_domain, self.service, self.user_id, post_id, headers, self.logger, cookies_dict=cookies)
                     if full_data:
                         final_post_data = full_data
@@ -1935,7 +1993,9 @@ class DownloadThread(QThread):
                  project_root_dir=None,
                  processed_post_ids=None,
                  start_offset=0,
-                 fetch_first=False): 
+                 fetch_first=False,
+                 skip_file_size_mb=None
+                 ): 
         super().__init__()
         self.api_url_input = api_url_input
         self.output_dir = output_dir
@@ -2002,6 +2062,7 @@ class DownloadThread(QThread):
         self.processed_post_ids_set = set(processed_post_ids) if processed_post_ids is not None else set() 
         self.start_offset = start_offset 
         self.fetch_first = fetch_first
+        self.skip_file_size_mb = skip_file_size_mb
 
         if self.compress_images and Image is None:
             self.logger("⚠️ Image compression disabled: Pillow library not found (DownloadThread).")
@@ -2122,6 +2183,7 @@ class DownloadThread(QThread):
                         'single_pdf_mode': self.single_pdf_mode,
                         'multipart_parts_count': self.multipart_parts_count, 
                         'multipart_min_size_mb': self.multipart_min_size_mb, 
+                        'skip_file_size_mb': self.skip_file_size_mb, 
                         'project_root_dir': self.project_root_dir,
                     }
 

src/ui/main_window.py

@@ -281,7 +281,7 @@ class DownloaderApp (QWidget ):
         self.download_location_label_widget = None
         self.remove_from_filename_label_widget = None
         self.skip_words_label_widget = None
-        self.setWindowTitle("Kemono Downloader v6.3.1")
+        self.setWindowTitle("Kemono Downloader v6.4.2")
         setup_ui(self)
         self._connect_signals()
         self.log_signal.emit("ℹ️ Local API server functionality has been removed.")
@@ -3354,7 +3354,8 @@ class DownloaderApp (QWidget ):
                             'pause_event': self.pause_event, 'cancellation_event': self.cancellation_event,
                             'downloaded_files': self.downloaded_files, 'downloaded_file_hashes': self.downloaded_file_hashes,
                             'downloaded_files_lock': self.downloaded_files_lock, 'downloaded_file_hashes_lock': self.downloaded_file_hashes_lock,
-                            'skip_words_list': [word.strip().lower() for word in self.skip_words_input.text().strip().split(',') if word.strip()],
+                            'skip_words_list': [part.strip().lower() for part in self.skip_words_input.text().strip().split(',') if part.strip() and not part.strip().startswith('[')],
+                            'skip_file_size_mb': next((int(re.search(r'\[(\d+)\]', part).group(1)) for part in self.skip_words_input.text().strip().split(',') if re.fullmatch(r'\[\d+\]', part.strip())), None),
                             'skip_words_scope': self.get_skip_words_scope(), 'char_filter_scope': self.get_char_filter_scope(),
                             'remove_from_filename_words_list': [word.strip() for word in self.remove_from_filename_input.text().strip().split(',') if word.strip()],
                             'scan_content_for_images': self.scan_content_images_checkbox.isChecked(),
@@ -3523,8 +3524,19 @@ class DownloaderApp (QWidget ):
                     self.thread_count_input.selectAll()
                     return False
 
-        raw_skip_words = self.skip_words_input.text().strip()
+        raw_skip_words_text = self.skip_words_input.text().strip()
-        skip_words_list = [word.strip().lower() for word in raw_skip_words.split(',') if word.strip()]
+        skip_words_parts = [part.strip() for part in raw_skip_words_text.split(',') if part.strip()]
+        skip_words_list = []
+        skip_file_size_mb = None
+        size_pattern = re.compile(r'\[(\d+)\]')
+
+        for part in skip_words_parts:
+            match = size_pattern.fullmatch(part)
+            if match:
+                skip_file_size_mb = int(match.group(1))
+                self.log_signal.emit(f"ℹ️ File size skip rule found: Will skip files smaller than {skip_file_size_mb} MB.")
+            else:
+                skip_words_list.append(part.lower())
 
         raw_remove_filename_words = self.remove_from_filename_input.text().strip() if hasattr(self, 'remove_from_filename_input') else ""
         allow_multipart = self.allow_multipart_download_setting
@@ -3891,6 +3903,7 @@ class DownloaderApp (QWidget ):
             'downloaded_file_hashes': self.downloaded_file_hashes,
             'downloaded_file_hashes_lock': self.downloaded_file_hashes_lock,
             'skip_words_list': skip_words_list,
+            'skip_file_size_mb': skip_file_size_mb,
             'skip_words_scope': current_skip_words_scope,
             'remove_from_filename_words_list': remove_from_filename_words_list,
             'char_filter_scope': current_char_filter_scope,
@@ -5484,7 +5497,8 @@ class DownloaderApp (QWidget ):
             'downloaded_files_lock': self.downloaded_files_lock,
             'downloaded_file_hashes_lock': self.downloaded_file_hashes_lock,
             'dynamic_character_filter_holder': self.dynamic_character_filter_holder,
-            'skip_words_list': [word.strip().lower() for word in self.skip_words_input.text().strip().split(',') if word.strip()],
+            'skip_words_list': [part.strip().lower() for part in self.skip_words_input.text().strip().split(',') if part.strip() and not part.strip().startswith('[')],
+            'skip_file_size_mb': next((int(re.search(r'\[(\d+)\]', part).group(1)) for part in self.skip_words_input.text().strip().split(',') if re.fullmatch(r'\[\d+\]', part.strip())), None),
             'skip_words_scope': self.get_skip_words_scope(), 
             'show_external_links': self.external_links_checkbox.isChecked(),
             'extract_links_only': self.radio_only_links.isChecked(),