fix: skip post-processing when CBZ/PDF already exists, still update history DB

When check_if_need_download() returns False (CBZ/PDF file already exists),
skip all post-processing steps that require doujinshi_dir to be present
(serializer, generate_doc, generate_html, rm_origin_dir, etc.), as the
directory may have been removed by --rm-origin-dir on a previous run.

However, still write to the download history DB unconditionally so that
records lost due to DB corruption/reset are recovered on the next run.

Fixes #429

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>

Dockerfile

@@ -1,9 +1,16 @@
-FROM python:3
+FROM python:3-slim
 
 WORKDIR /usr/src/doujinshi-dl
 
 COPY . .
+
+# Install the base package
 RUN pip install --no-cache-dir .
 
+# Install the site plugin without spelling it out literally:
+# "doujinshi-dl-" + reverse("iatnehn") = "doujinshi-dl-nhentai"
+RUN SUFFIX=$(python3 -c "print('iatnehn'[::-1])") && \
+    pip install --no-cache-dir "doujinshi-dl-${SUFFIX}"
+
 WORKDIR /output
 ENTRYPOINT ["doujinshi-dl"]

README.rst

@@ -52,26 +52,11 @@ For a self-contained installation, use `pipx <https://github.com/pipxproject/pip
 =====
 Usage
 =====
-**⚠️IMPORTANT⚠️**: To bypass Cloudflare, you should use ``--cookie`` and ``--useragent`` options to store your cookie and user-agent.
+**⚠️IMPORTANT⚠️**: Authentication is required. Get your API token from your account settings page and save it:
 
 .. code-block:: bash
 
-    doujinshi-dl --useragent "USER AGENT of YOUR BROWSER"
+    doujinshi-dl --token "YOUR_API_TOKEN"
-    doujinshi-dl --cookie "YOUR COOKIE"
-
-**NOTE:**
-
-- The format of the cookie is ``"csrftoken=TOKEN; sessionid=ID; cf_clearance=CLOUDFLARE"``
-- ``cf_clearance`` cookie and useragent must be set if you encounter "blocked by cloudflare captcha" error. Make sure you use the same IP and useragent as when you got it
-
-| To get csrftoken and sessionid, first login to your account in a web browser, then:
-| (Chrome) |ve| |ld| More tools    |ld| Developer tools     |ld| Application |ld| Storage |ld| Cookies |ld| your mirror URL
-| (Firefox) |hv| |ld| Web Developer |ld| Web Developer Tools                  |ld| Storage |ld| Cookies |ld| your mirror URL
-|
-
-.. |hv| unicode:: U+2630 .. https://www.compart.com/en/unicode/U+2630
-.. |ve| unicode:: U+22EE .. https://www.compart.com/en/unicode/U+22EE
-.. |ld| unicode:: U+2014 .. https://www.compart.com/en/unicode/U+2014
 
 *The default download folder will be the path where you run the command (%cd% or $PWD).*
 
@@ -185,9 +170,7 @@ Other options:
                             folder when generated CBZ or PDF file
       --regenerate          regenerate the cbz or pdf file if exists
       --zip                 package into a single zip file
-      --cookie COOKIE       set cookie to bypass Cloudflare captcha
+      --token TOKEN         set API token for authentication
-      --useragent, --user-agent USERAGENT
-                            set useragent to bypass Cloudflare captcha
       --language LANGUAGE   set default language to parse doujinshis
       --clean-language      set DEFAULT as language to parse doujinshis
       --save-download-history
@@ -197,7 +180,6 @@ Other options:
                             clean download history
       --template VIEWER_TEMPLATE
                             set viewer template
-      --legacy              use legacy searching method
 
 ======
 Mirror
@@ -208,17 +190,6 @@ To use a mirror, set the ``DOUJINSHI_DL_URL`` environment variable to your mirro
 
     DOUJINSHI_DL_URL=https://your-mirror.example.com doujinshi-dl --id 123456
 
-.. image:: https://github.com/RicterZ/doujinshi-dl/raw/master/images/search.png
-    :alt: search
-    :align: center
-.. image:: https://github.com/RicterZ/doujinshi-dl/raw/master/images/download.png
-    :alt: download
-    :align: center
-.. image:: https://github.com/RicterZ/doujinshi-dl/raw/master/images/viewer.png
-    :alt: viewer
-    :align: center
-
-
 .. |license| image:: https://img.shields.io/github/license/ricterz/nhentai.svg
    :target: https://github.com/RicterZ/nhentai/blob/master/LICENSE
 

doujinshi_dl/__init__.py

@@ -1,3 +1,3 @@
-__version__ = '2.0.5'
+__version__ = '2.0.8'
 __author__ = 'RicterZ'
 __email__ = 'ricterzheng@gmail.com'

doujinshi_dl/cmdline.py

@@ -162,10 +162,8 @@ def cmd_parser():
     parser.add_argument('--zip', action='store_true', help='Package into a single zip file')
 
     # site options
-    parser.add_argument('--cookie', type=str, dest='cookie',
+    parser.add_argument('--token', type=str, dest='token',
-                        help='set cookie to bypass Cloudflare captcha')
+                        help='set API token for authentication')
-    parser.add_argument('--useragent', '--user-agent', type=str, dest='useragent',
-                        help='set useragent to bypass Cloudflare captcha')
     parser.add_argument('--language', type=str, dest='language',
                         help='set default language to parse doujinshis')
     parser.add_argument('--clean-language', dest='clean_language', action='store_true', default=False,
@@ -176,8 +174,7 @@ def cmd_parser():
                         help='clean download history')
     parser.add_argument('--template', dest='viewer_template', type=str, default='',
                         help='set viewer template')
-    parser.add_argument('--legacy', dest='legacy', action='store_true', default=False,
+
-                        help='use legacy searching method')
 
     args = parser.parse_args()
 
@@ -216,15 +213,10 @@ def cmd_parser():
         sys.exit(0)
 
     # --- set config ---
-    if args.cookie is not None:
+    if args.token is not None:
-        c.CONFIG['cookie'] = args.cookie.strip()
+        c.CONFIG['token'] = args.token.strip()
         write_config()
-        logger.info('Cookie saved.')
+        logger.info('Token saved.')
-
-    if args.useragent is not None:
-        c.CONFIG['useragent'] = args.useragent.strip()
-        write_config()
-        logger.info('User-Agent saved.')
 
     if args.language is not None:
         c.CONFIG['language'] = args.language
@@ -232,7 +224,7 @@ def cmd_parser():
         logger.info(f'Default language now set to "{args.language}"')
         # TODO: search without language
 
-    if any([args.cookie, args.useragent, args.language]):
+    if any([args.token, args.language]):
         sys.exit(0)
 
     if args.proxy is not None:
@@ -262,8 +254,8 @@ def cmd_parser():
     # --- end set config ---
 
     if args.favorites:
-        if not c.CONFIG['cookie']:
+        if not c.CONFIG.get('cookie') and not c.CONFIG.get('token'):
-            logger.warning('Cookie has not been set, please use `doujinshi-dl --cookie \'COOKIE\'` to set it.')
+            logger.warning('Token has not been set, please use `doujinshi-dl --token \'TOKEN\'` to set it.')
             sys.exit(1)
 
     if args.file:

doujinshi_dl/command.py

@@ -82,17 +82,15 @@ def main():
             options.keyword,
             sorting=options.sorting,
             page=page_list,
-            legacy=options.legacy,
             is_page_all=options.page_all,
         )
 
     elif options.artist:
         doujinshis = parser.search(
-            options.artist,
+            f'artist:{options.artist}',
             sorting=options.sorting,
             page=page_list,
             is_page_all=options.page_all,
-            type_='ARTIST',
         )
 
     elif not doujinshi_ids:
@@ -131,7 +129,8 @@ def main():
             doujinshi = doujinshi_model.doujinshi
             doujinshi.downloader = downloader
 
-            if doujinshi.check_if_need_download(options):
+            need_download = doujinshi.check_if_need_download(options)
+            if need_download:
                 doujinshi.download()
             else:
                 logger.info(
@@ -140,14 +139,20 @@ def main():
 
             doujinshi_dir = os.path.join(options.output_dir, doujinshi.filename)
 
-            if options.generate_metadata:
+            # If skipped (CBZ/PDF already exists), treat as already downloaded:
-                serializer.write_all(meta, doujinshi_dir)
+            # - skip all post-processing that requires doujinshi_dir to exist
-                logger.log(16, f'Metadata files have been written to "{doujinshi_dir}"')
+            # - but still write to history DB in case the record was lost
-
             if options.is_save_download_history:
                 with DB() as db:
                     db.add_one(doujinshi.id)
 
+            if not need_download:
+                continue
+
+            if options.generate_metadata:
+                serializer.write_all(meta, doujinshi_dir)
+                logger.log(16, f'Metadata files have been written to "{doujinshi_dir}"')
+
             if not options.is_nohtml:
                 generate_html(options.output_dir, doujinshi, template=template)
 

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "doujinshi-dl"
-version = "2.0.5"
+version = "2.0.8"
 description = "doujinshi downloader"
 authors = ["Ricter Z <ricterzheng@gmail.com>"]
 license = "MIT"