improve download logic #343

nhentai/command.py

@@ -88,10 +88,10 @@ def main():
             if not options.dryrun:
                 doujinshi.downloader = downloader
 
-                result = doujinshi.download(skip_exists=not options.regenerate)
+                if doujinshi.check_if_need_download(options):
-                # Already downloaded; continue on with the other doujins.
+                    doujinshi.download()
-                if not result:
+                else:
-                    continue
+                    logger.info(f'Skip download doujinshi because a PDF/CBZ file exists of doujinshi {doujinshi.name}')
 
             if options.generate_metadata:
                 generate_metadata_file(options.output_dir, doujinshi)

nhentai/doujinshi.py

@@ -72,14 +72,34 @@ class Doujinshi(object):
     def show(self):
         logger.info(f'Print doujinshi information of {self.id}\n{tabulate(self.table)}')
 
-    def download(self, skip_exists=True):
+    def check_if_need_download(self, options):
-        logger.info(f'Starting to download doujinshi: {self.name}')
-
         base_path = os.path.join(self.downloader.path, self.filename)
-        if (os.path.exists(base_path + '.pdf') or os.path.exists(base_path + '.cbz')) and skip_exists:
+
-            logger.info(f'Skip download doujinshi because a PDF/CBZ file exists of doujinshi {self.name}')
+        # doujinshi directory is not exist, we need to download definitely
+        if not (os.path.exists(base_path) and os.path.isdir(base_path)):
+            return True
+
+        # regenerate, we need to re-download from nhentai
+        if options.regenerate:
+            return True
+
+        if options.is_pdf:
+            file_ext = 'pdf'
+        elif options.is_cbz:
+            file_ext = 'cbz'
+        else:
+            # re-download
+            return True
+
+        # pdf or cbz file exists, we needn't to re-download it
+        if os.path.exists(f'{base_path}.{file_ext}') or os.path.exists(f'{base_path}/{self.filename}.{file_ext}'):
             return False
 
+        # fallback
+        return True
+
+    def download(self):
+        logger.info(f'Starting to download doujinshi: {self.name}')
         if self.downloader:
             download_queue = []
             if len(self.ext) != self.pages:

nhentai/utils.py

@@ -191,7 +191,7 @@ def generate_doc(file_type='', output_dir='.', doujinshi_obj=None, rm_origin_dir
     doujinshi_dir, filename = parse_doujinshi_obj(output_dir, doujinshi_obj, file_type)
 
     if os.path.exists(f'{doujinshi_dir}.{file_type}') and not regenerate:
-        logger.info(f'Skipped download: {doujinshi_dir}.{file_type} already exists')
+        logger.info(f'Skipped {file_type} file generation: {doujinshi_dir}.{file_type} already exists')
         return
 
     if file_type == 'cbz':