0.5.25, fix #393

Update installation instructions in README.rst

README.rst

@@ -22,7 +22,7 @@ From Github:
 
     git clone https://github.com/RicterZ/nhentai
     cd nhentai
-    python setup.py install
+    pip install --no-cache-dir .
 
 Build Docker container:
 
@@ -136,6 +136,8 @@ Format output doujinshi folder name:
 .. code-block:: bash
 
     nhentai --id 261100 --format '[%i]%s'
+    # for Windows
+    nhentai --id 261100 --format "[%%i]%%s"
 
 Supported doujinshi folder formatter:
 
@@ -148,6 +150,7 @@ Supported doujinshi folder formatter:
 - %p: Doujinshi pretty name
 - %ag: Doujinshi authors name or groups name
 
+Note: for Windows operation system, please use double "%", such as "%%i".
 
 Other options:
 

nhentai/__init__.py

@@ -1,3 +1,3 @@
-__version__ = '0.5.20'
+__version__ = '0.5.25'
 __author__ = 'RicterZ'
 __email__ = 'ricterzheng@gmail.com'

nhentai/cmdline.py

@@ -131,6 +131,8 @@ def cmd_parser():
                       help='generate a metadata file in doujinshi format')
     parser.add_option('--regenerate', dest='regenerate', action='store_true', default=False,
                       help='regenerate the cbz or pdf file if exists')
+    parser.add_option('--no-metadata', dest='no_metadata', action='store_true', default=False,
+                      help='don\'t generate metadata json file in doujinshi output path')
 
     # nhentai options
     parser.add_option('--cookie', type='str', dest='cookie', action='store',
@@ -169,22 +171,24 @@ def cmd_parser():
 
     # --- set config ---
     if args.cookie is not None:
-        constant.CONFIG['cookie'] = args.cookie
+        constant.CONFIG['cookie'] = args.cookie.strip()
         write_config()
         logger.info('Cookie saved.')
-        sys.exit(0)
+
-    elif args.useragent is not None:
+    if args.useragent is not None:
-        constant.CONFIG['useragent'] = args.useragent
+        constant.CONFIG['useragent'] = args.useragent.strip()
         write_config()
         logger.info('User-Agent saved.')
-        sys.exit(0)
+
-    elif args.language is not None:
+    if args.language is not None:
         constant.CONFIG['language'] = args.language
         write_config()
         logger.info(f'Default language now set to "{args.language}"')
-        sys.exit(0)
         # TODO: search without language
 
+    if any([args.cookie, args.useragent, args.language]):
+        sys.exit(0)
+
     if args.proxy is not None:
         proxy_url = urlparse(args.proxy)
         if not args.proxy == '' and proxy_url.scheme not in ('http', 'https', 'socks5', 'socks5h',

nhentai/command.py

@@ -4,8 +4,6 @@ import shutil
 import sys
 import signal
 import platform
-import urllib
-
 import urllib3.exceptions
 
 from nhentai import constant
@@ -51,6 +49,9 @@ def main():
 
     page_list = paging(options.page)
 
+    if options.retry:
+        constant.RETRY_TIMES = int(options.retry)
+
     if options.favorites:
         if not options.is_download:
             logger.warning('You do not specify --download option')
@@ -86,7 +87,7 @@ def main():
     if not options.is_show:
         downloader = Downloader(path=options.output_dir, threads=options.threads,
                                 timeout=options.timeout, delay=options.delay,
-                                retry=options.retry, exit_on_fail=options.exit_on_fail,
+                                exit_on_fail=options.exit_on_fail,
                                 no_filename_padding=options.no_filename_padding)
 
         for doujinshi_id in doujinshi_ids:
@@ -115,6 +116,9 @@ def main():
             if not options.is_nohtml:
                 generate_html(options.output_dir, doujinshi, template=constant.CONFIG['template'])
 
+            if not options.no_metadata:
+                generate_doc('json', options.output_dir, doujinshi, options.regenerate)
+
             if options.is_cbz:
                 generate_doc('cbz', options.output_dir, doujinshi, options.regenerate)
 

nhentai/constant.py

@@ -37,6 +37,8 @@ FAV_URL = f'{BASE_URL}/favorites/'
 
 PATH_SEPARATOR = os.path.sep
 
+RETRY_TIMES = 3
+
 
 IMAGE_URL = f'{urlparse(BASE_URL).scheme}://i1.{urlparse(BASE_URL).hostname}/galleries'
 IMAGE_URL_MIRRORS = [

nhentai/downloader.py

@@ -34,13 +34,12 @@ def download_callback(result):
 
 
 class Downloader(Singleton):
-    def __init__(self, path='', threads=5, timeout=30, delay=0, retry=3, exit_on_fail=False,
+    def __init__(self, path='', threads=5, timeout=30, delay=0, exit_on_fail=False,
                  no_filename_padding=False):
         self.threads = threads
         self.path = str(path)
         self.timeout = timeout
         self.delay = delay
-        self.retry = retry
         self.exit_on_fail = exit_on_fail
         self.folder = None
         self.semaphore = None
@@ -101,7 +100,7 @@ class Downloader(Singleton):
                 return -1, url
 
         except (httpx.HTTPStatusError, httpx.TimeoutException, httpx.ConnectError) as e:
-            if retried < self.retry:
+            if retried < constant.RETRY_TIMES:
                 logger.warning(f'Download {filename} failed, retrying({retried + 1}) times...')
                 return await self.download(
                     url=url,
@@ -111,7 +110,7 @@ class Downloader(Singleton):
                     proxy=proxy,
                 )
             else:
-                logger.warning(f'Download {filename} failed with {self.retry} times retried, skipped')
+                logger.warning(f'Download {filename} failed with {constant.RETRY_TIMES} times retried, skipped')
                 return -2, url
 
         except NHentaiImageNotExistException as e:

nhentai/parser.py

@@ -92,13 +92,27 @@ def favorites_parser(page=None):
         page_range_list = range(1, pages + 1)
 
     for page in page_range_list:
-        try:
+        logger.info(f'Getting doujinshi ids of page {page}')
-            logger.info(f'Getting doujinshi ids of page {page}')
-            resp = request('get', f'{constant.FAV_URL}?page={page}').content
 
-            result.extend(_get_title_and_id(resp))
+        i = 0
-        except Exception as e:
+        while i <= constant.RETRY_TIMES + 1:
-            logger.error(f'Error: {e}, continue')
+            i += 1
+            if i > 3:
+                logger.error(f'Failed to get favorites at page {page} after 3 times retried, skipped')
+                break
+
+            try:
+                resp = request('get', f'{constant.FAV_URL}?page={page}').content
+                temp_result = _get_title_and_id(resp)
+                if not temp_result:
+                    logger.warning(f'Failed to get favorites at page {page}, retrying ({i} times) ...')
+                    continue
+                else:
+                    result.extend(temp_result)
+                    break
+
+            except Exception as e:
+                logger.warning(f'Error: {e}, retrying ({i} times) ...')
 
     return result
 
@@ -141,17 +155,19 @@ def doujinshi_parser(id_, counter=0):
     title = doujinshi_info.find('h1').text
     pretty_name = doujinshi_info.find('h1').find('span', attrs={'class': 'pretty'}).text
     subtitle = doujinshi_info.find('h2')
-    favorite_counts = doujinshi_info.find('span', class_='nobold').find('span', class_='count')
+    favorite_counts = doujinshi_info.find('span', class_='nobold').text.strip('(').strip(')')
 
     doujinshi['name'] = title
     doujinshi['pretty_name'] = pretty_name
     doujinshi['subtitle'] = subtitle.text if subtitle else ''
-    doujinshi['favorite_counts'] = int(favorite_counts.text.strip()) if favorite_counts else 0
+    doujinshi['favorite_counts'] = int(favorite_counts) if favorite_counts and favorite_counts.isdigit() else 0
 
     doujinshi_cover = html.find('div', attrs={'id': 'cover'})
     # img_id = re.search('/galleries/([0-9]+)/cover.(jpg|png|gif|webp)$',
     #                   doujinshi_cover.a.img.attrs['data-src'])
-    img_id = re.search(r'/galleries/(\d+)/cover\.\w+$', doujinshi_cover.a.img.attrs['data-src'])
+
+    # fix cover.webp.webp
+    img_id = re.search(r'/galleries/(\d+)/cover(.webp)?\.\w+$', doujinshi_cover.a.img.attrs['data-src'])
 
     ext = []
     for i in html.find_all('div', attrs={'class': 'thumb-container'}):
@@ -261,7 +277,7 @@ def search_parser(keyword, sorting, page, is_page_all=False):
         i = 0
 
         logger.info(f'Searching doujinshis using keywords "{keyword}" on page {p}{total}')
-        while i < 3:
+        while i < constant.RETRY_TIMES:
             try:
                 url = request('get', url=constant.SEARCH_URL, params={'query': keyword,
                                                                       'page': p, 'sort': sorting}).url

nhentai/serializer.py

@@ -2,12 +2,11 @@
 import json
 import os
 
-from nhentai.constant import PATH_SEPARATOR
+from nhentai.constant import PATH_SEPARATOR, LANGUAGE_ISO
 from xml.sax.saxutils import escape
-from nhentai.constant import LANGUAGE_ISO
 
 
-def serialize_json(doujinshi, output_dir):
+def serialize_json(doujinshi, output_dir: str):
     metadata = {'title': doujinshi.name,
                 'subtitle': doujinshi.info.subtitle}
     if doujinshi.info.favorite_counts:

nhentai/utils.py

@@ -16,7 +16,7 @@ from requests.structures import CaseInsensitiveDict
 from nhentai import constant
 from nhentai.constant import PATH_SEPARATOR
 from nhentai.logger import logger
-from nhentai.serializer import serialize_json, serialize_comic_xml, set_js_database
+from nhentai.serializer import serialize_comic_xml, serialize_json, set_js_database
 
 MAX_FIELD_LENGTH = 100
 EXTENSIONS = ('.png', '.jpg', '.jpeg', '.gif', '.webp')
@@ -142,7 +142,7 @@ def generate_html(output_dir='.', doujinshi_obj=None, template='default'):
     js = readfile(f'viewer/{template}/scripts.js')
 
     if doujinshi_obj is not None:
-        serialize_json(doujinshi_obj, doujinshi_dir)
+        # serialize_json(doujinshi_obj, doujinshi_dir)
         name = doujinshi_obj.name
     else:
         name = {'title': 'nHentai HTML Viewer'}
@@ -274,6 +274,9 @@ def generate_doc(file_type='', output_dir='.', doujinshi_obj=None, regenerate=Fa
         except ImportError:
             logger.error("Please install img2pdf package by using pip.")
 
+    elif file_type == 'json':
+        serialize_json(doujinshi_obj, doujinshi_dir)
+
 
 def format_filename(s, length=MAX_FIELD_LENGTH, _truncate_only=False):
     """

nhentai/viewer/default/scripts.js

@@ -49,8 +49,8 @@ document.onkeypress = event => {
     switch (event.key.toLowerCase()) {
         // Previous Image
         case 'w':
-	   scrollBy(0, -40);
+            scrollBy(0, -40);
-	   break;
+            break;
         case 'a':
             changePage(currentPage - 1);
             break;
@@ -61,7 +61,7 @@ document.onkeypress = event => {
         // Next Image
         case ' ':
         case 's':
-	    scrollBy(0, 40);
+            scrollBy(0, 40);
             break;
         case 'd':
             changePage(currentPage + 1);
@@ -75,11 +75,13 @@ document.onkeydown = event =>{
             changePage(currentPage - 1);
             break;
         case 38: //up
+            changePage(currentPage - 1);
             break;
         case 39: //right
             changePage(currentPage + 1);
             break;
         case 40: //down
+            changePage(currentPage + 1);
             break;
     }
 };

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "nhentai"
-version = "0.5.20"
+version = "0.5.25"
 description = "nhentai doujinshi downloader"
 authors = ["Ricter Z <ricterzheng@gmail.com>"]
 license = "MIT"