fix #29 0.2.16

Max retries + misc. language fixes

.gitignore

@@ -4,4 +4,4 @@ build
 dist/
 *.egg-info
 .python-version
-
+.DS_Store

.travis.yml

@@ -4,10 +4,9 @@ os:
 language: python
 python:
     - 2.7
-    - 2.6
-    - 3.3
+    - 3.6
+    - 3.5
     - 3.4
-    - 3.5.2
 
 install:
     - python setup.py install
@@ -15,3 +14,5 @@ install:
 script:
     - NHENTAI=https://nhentai.net nhentai --search umaru
     - NHENTAI=https://nhentai.net nhentai --id=152503,146134 -t 10 --output=/tmp/
+    - NHENTAI=https://nhentai.net nhentai -l nhentai_test:nhentai --output=/tmp/
+    - NHENTAI=https://nhentai.net nhentai --tag lolicon

MANIFEST.in

@@ -1,2 +1,5 @@
 include README.md
 include requirements.txt
+include nhentai/viewer/index.html
+include nhentai/viewer/styles.css
+include nhentai/viewer/scripts.js

README.md

@@ -7,46 +7,69 @@ nhentai
     |_| |_|_| |_|\___|_| |_|\__\__,_|_|
 
 あなたも変態。 いいね?  
-[![Build Status](https://travis-ci.org/RicterZ/nhentai.svg?branch=master)](https://travis-ci.org/RicterZ/nhentai)  
+[![Build Status](https://travis-ci.org/RicterZ/nhentai.svg?branch=master)](https://travis-ci.org/RicterZ/nhentai) ![nhentai PyPI Downloads](https://pypistats.com/badge/nhentai.svg)  
 
-🎉🎉 nhentai 现在支持 Windows 啦！
 
-由于 [http://nhentai.net](http://nhentai.net) 下载下来的种子速度很慢，而且官方也提供在线观看本子的功能，所以可以利用本脚本下载本子。
-### 安装
+nHentai is a CLI tool for downloading doujinshi from [nhentai.net](http://nhentai.net).
+
+### Installation
 
     git clone https://github.com/RicterZ/nhentai
     cd nhentai
     python setup.py install
     
+### Gentoo
 
-### 用法
-+ 下载指定 id 列表的本子：
+    layman -fa glicOne
+    sudo emerge net-misc/nhentai
 
+### Usage
+Download specified doujinshi:
+```bash
+nhentai --id=123855,123866
+```
 
-    nhentai --id=123855,123866
+Search a keyword and download the first page:
+```bash
+nhentai --search="tomori" --page=1 --download
+```
 
+Download your favourite doujinshi (login required):
+```bash
+nhentai --login "username:password" --download
+```
 
-+ 下载某关键词第一页的本子（不推荐）：
+Download by tag name:
+```bash
+nhentai --tag lolicon --download
+```
 
+### Options
 
-    nhentai --search="tomori" --page=1 --download
++ `-t, --thread`: Download threads, max: 10  
++ `--output`:Output dir of saving doujinshi  
++ `--tag`:Download by tag name  
++ `--timeout`: Timeout of downloading each image   
++ `--proxy`: Use proxy, example: http://127.0.0.1:8080/  
++ `--login`: username:password pair of your nhentai account  
++ `--nohtml`: Do not generate HTML  
++ `--cbz`: Generate Comic Book CBZ File  
 
+### nHentai Mirror
+If you want to use a mirror, you should set up a reverse proxy of `nhentai.net` and `i.nhentai.net`.
+For example:
 
-`-t, --thread`：指定下载的线程数，最多为 10 线程。  
-`--path`：指定下载文件的输出路径，默认为当前目录。  
-`--timeout`：指定下载图片的超时时间，默认为 30 秒。  
-`--proxy`：指定下载的代理，例如: http://127.0.0.1:8080/
-
-### 自建 nhentai 镜像
-如果想用自建镜像下载 nhentai 的本子，需要搭建 nhentai.net 和 i.nhentai.net 的反向代理。  
-例如用 h.loli.club 来做反向代理的话，需要 h.loli.club 反代 nhentai.net，i.h.loli.club 反带 i.nhentai.net。  
-然后利用环境变量来下载：  
-
-    NHENTAI=http://h.loli.club nhentai --id 123456
+    i.h.loli.club -> i.nhentai.net
+    h.loli.club -> nhentai.net
 
+Set `NHENTAI` env var to your nhentai mirror.
+```bash
+NHENTAI=http://h.loli.club nhentai --id 123456
+```
 
 ![](./images/search.png)  
 ![](./images/download.png)  
+![](./images/viewer.png)  
 
 ### License  
 MIT

nhentai/__init__.py

@@ -1,3 +1,3 @@
-__version__ = '0.2.4'
-__author__ = 'Ricter'
+__version__ = '0.2.16'
+__author__ = 'RicterZ'
 __email__ = 'ricterzheng@gmail.com'

nhentai/cmdline.py

@@ -2,31 +2,36 @@
 from __future__ import print_function
 import sys
 from optparse import OptionParser
+from nhentai import __version__
 try:
     from itertools import ifilter as filter
 except ImportError:
     pass
 
 import nhentai.constant as constant
-from nhentai.utils import urlparse
+from nhentai.utils import urlparse, generate_html
 from nhentai.logger import logger
 
 try:
-    reload(sys)
-    sys.setdefaultencoding(sys.stdin.encoding)
+    if sys.version_info < (3, 0, 0):
+        import codecs
+        import locale
+        sys.stdout = codecs.getwriter(locale.getpreferredencoding())(sys.stdout)
+        sys.stderr = codecs.getwriter(locale.getpreferredencoding())(sys.stderr)
+
 except NameError:
     # python3
     pass
 
 
 def banner():
-    logger.info(u'''nHentai: あなたも変態。 いいね?
+    logger.info(u'''nHentai ver %s: あなたも変態。 いいね?
        _   _            _        _
  _ __ | | | | ___ _ __ | |_ __ _(_)
 | '_ \| |_| |/ _ \ '_ \| __/ _` | |
 | | | |  _  |  __/ | | | || (_| | |
 |_| |_|_| |_|\___|_| |_|\__\__,_|_|
-''')
+''' % __version__)
 
 
 def cmd_parser():
@@ -34,21 +39,35 @@ def cmd_parser():
                           '\n  NHENTAI=http://h.loli.club nhentai --id [ID ...]'
                           '\n\nEnvironment Variable:\n'
                           '  NHENTAI                 nhentai mirror url')
-    parser.add_option('--download', dest='is_download', action='store_true', help='download doujinshi (for search result)')
+    parser.add_option('--download', dest='is_download', action='store_true',
+                      help='download doujinshi (for search results)')
     parser.add_option('--show-info', dest='is_show', action='store_true', help='just show the doujinshi information')
     parser.add_option('--id', type='string', dest='id', action='store', help='doujinshi ids set, e.g. 1,2,3')
     parser.add_option('--search', type='string', dest='keyword', action='store', help='search doujinshi by keyword')
     parser.add_option('--page', type='int', dest='page', action='store', default=1,
-                      help='page number of search result')
-    parser.add_option('--tags', type='string', dest='tags', action='store', help='download doujinshi by tags')
+                      help='page number of search results')
+    parser.add_option('--tag', type='string', dest='tag', action='store', help='download doujinshi by tag')
+    parser.add_option('--max-page', type='int', dest='max_page', action='store', default=1,
+                      help='The max page when recursive download tagged doujinshi')
     parser.add_option('--output', type='string', dest='output_dir', action='store', default='',
                       help='output dir')
     parser.add_option('--threads', '-t', type='int', dest='threads', action='store', default=5,
-                      help='thread count of download doujinshi')
+                      help='thread count for downloading doujinshi')
     parser.add_option('--timeout', type='int', dest='timeout', action='store', default=30,
-                      help='timeout of download doujinshi')
+                      help='timeout for downloading doujinshi')
     parser.add_option('--proxy', type='string', dest='proxy', action='store', default='',
-                      help='use proxy, example: http://127.0.0.1:1080')
+                      help='uses a proxy, for example: http://127.0.0.1:1080')
+    parser.add_option('--html', dest='html_viewer', action='store_true',
+                      help='generate a html viewer at current directory')
+
+    parser.add_option('--login', '-l', type='str', dest='login', action='store',
+                      help='username:password pair of nhentai account')
+
+    parser.add_option('--nohtml', dest='is_nohtml', action='store_true',
+                      help='Don\'t generate HTML')
+
+    parser.add_option('--cbz', dest='is_cbz', action='store_true',
+                      help='Generate Comic Book CBZ File')                      
                       
     try:
         sys.argv = list(map(lambda x: unicode(x.decode(sys.stdin.encoding)), sys.argv))
@@ -59,35 +78,46 @@ def cmd_parser():
 
     args, _ = parser.parse_args(sys.argv[1:])
 
-    if args.tags:
-        logger.warning('`--tags` is under construction')
+    if args.html_viewer:
+        generate_html()
         exit(0)
 
+    if args.login:
+        try:
+            _, _ = args.login.split(':', 1)
+        except ValueError:
+            logger.error('Invalid `username:password` pair.')
+            exit(1)
+
+        if not args.is_download:
+            logger.warning('YOU DO NOT SPECIFY `--download` OPTION !!!')
+
     if args.id:
         _ = map(lambda id: id.strip(), args.id.split(','))
-        args.id = set(map(int, filter(lambda id: id.isdigit(), _)))
+        args.id = set(map(int, filter(lambda id_: id_.isdigit(), _)))
 
-    if (args.is_download or args.is_show) and not args.id and not args.keyword:
+    if (args.is_download or args.is_show) and not args.id and not args.keyword and \
+            not args.login and not args.tag:
         logger.critical('Doujinshi id(s) are required for downloading')
         parser.print_help()
-        exit(0)
+        exit(1)
 
-    if not args.keyword and not args.id:
+    if not args.keyword and not args.id and not args.login and not args.tag:
         parser.print_help()
-        exit(0)
+        exit(1)
 
     if args.threads <= 0:
         args.threads = 1
 
     elif args.threads > 15:
         logger.critical('Maximum number of used threads is 15')
-        exit(0)
+        exit(1)
 
     if args.proxy:
         proxy_url = urlparse(args.proxy)
         if proxy_url.scheme not in ('http', 'https'):
             logger.error('Invalid protocol \'{0}\' of proxy, ignored'.format(proxy_url.scheme))
         else:
-            constant.PROXY = {proxy_url.scheme: args.proxy}
+            constant.PROXY = {'http': args.proxy, 'https': args.proxy}
 
     return args

nhentai/command.py

@@ -5,11 +5,12 @@ import signal
 import platform
 
 from nhentai.cmdline import cmd_parser, banner
-from nhentai.parser import doujinshi_parser, search_parser, print_doujinshi
+from nhentai.parser import doujinshi_parser, search_parser, print_doujinshi, login_parser, tag_guessing, tag_parser
 from nhentai.doujinshi import Doujinshi
 from nhentai.downloader import Downloader
 from nhentai.logger import logger
 from nhentai.constant import BASE_URL
+from nhentai.utils import generate_html, generate_cbz
 
 
 def main():
@@ -20,20 +21,35 @@ def main():
     doujinshi_ids = []
     doujinshi_list = []
 
+    import pdb; pdb.set_trace()
+
+    if options.login:
+        username, password = options.login.split(':', 1)
+        logger.info('Logging in to nhentai using credential pair \'%s:%s\'' % (username, '*' * len(password)))
+        for doujinshi_info in login_parser(username=username, password=password):
+            doujinshi_list.append(Doujinshi(**doujinshi_info))
+
+    if options.tag:
+        tag_id = tag_guessing(options.tag)
+        if tag_id:
+            doujinshis = tag_parser(tag_id, max_page=options.max_page)
+            print_doujinshi(doujinshis)
+            if options.is_download:
+                doujinshi_ids = map(lambda d: d['id'], doujinshis)
+
     if options.keyword:
         doujinshis = search_parser(options.keyword, options.page)
         print_doujinshi(doujinshis)
         if options.is_download:
             doujinshi_ids = map(lambda d: d['id'], doujinshis)
-    else:
+
+    if not doujinshi_ids:
         doujinshi_ids = options.id
 
     if doujinshi_ids:
-        for id in doujinshi_ids:
-            doujinshi_info = doujinshi_parser(id)
+        for id_ in doujinshi_ids:
+            doujinshi_info = doujinshi_parser(id_)
             doujinshi_list.append(Doujinshi(**doujinshi_info))
-    else:
-        exit(0)
 
     if not options.is_show:
         downloader = Downloader(path=options.output_dir,
@@ -42,9 +58,13 @@ def main():
         for doujinshi in doujinshi_list:
             doujinshi.downloader = downloader
             doujinshi.download()
+            if not options.is_nohtml and not options.is_cbz:
+                generate_html(options.output_dir, doujinshi)
+            elif options.is_cbz:
+                generate_cbz(options.output_dir, doujinshi)
 
         if not platform.system() == 'Windows':
-            logger.log(15, '🍺 All done.')
+            logger.log(15, '🍻 All done.')
         else:
             logger.log(15, 'All done.')
 
@@ -53,7 +73,7 @@ def main():
 
 
 def signal_handler(signal, frame):
-    logger.error('Ctrl-C signal received. Quit.')
+    logger.error('Ctrl-C signal received. Stopping...')
     exit(1)
 
 

nhentai/constant.py

@@ -5,8 +5,12 @@ from nhentai.utils import urlparse
 
 BASE_URL = os.getenv('NHENTAI', 'https://nhentai.net')
 
-DETAIL_URL = '%s/g' % BASE_URL
-SEARCH_URL = '%s/search/' % BASE_URL
+DETAIL_URL = '%s/api/gallery' % BASE_URL
+SEARCH_URL = '%s/api/galleries/search' % BASE_URL
+TAG_URL = '%s/tag' % BASE_URL
+TAG_API_URL = '%s/api/galleries/tagged' % BASE_URL
+LOGIN_URL = '%s/login/' % BASE_URL
+FAV_URL = '%s/favorites/' % BASE_URL
 
 u = urlparse(BASE_URL)
 IMAGE_URL = '%s://i.%s/galleries' % (u.scheme, u.hostname)

nhentai/doujinshi.py

@@ -5,6 +5,14 @@ from future.builtins import range
 
 from nhentai.constant import DETAIL_URL, IMAGE_URL
 from nhentai.logger import logger
+from nhentai.utils import format_filename
+
+
+EXT_MAP = {
+    'j': 'jpg',
+    'p': 'png',
+    'g': 'gif',
+}
 
 
 class DoujinshiInfo(dict):
@@ -19,7 +27,7 @@ class DoujinshiInfo(dict):
 
 
 class Doujinshi(object):
-    def __init__(self, name=None, id=None, img_id=None, ext='jpg', pages=0, **kwargs):
+    def __init__(self, name=None, id=None, img_id=None, ext='', pages=0, **kwargs):
         self.name = name
         self.id = id
         self.img_id = img_id
@@ -46,14 +54,15 @@ class Doujinshi(object):
         logger.info(u'Print doujinshi information of {0}\n{1}'.format(self.id, tabulate(table)))
 
     def download(self):
-        logger.info('Start download doujinshi: %s' % self.name)
+        logger.info('Starting to download doujinshi: %s' % self.name)
         if self.downloader:
             download_queue = []
-            for i in range(1, self.pages + 1):
-                download_queue.append('%s/%d/%d.%s' % (IMAGE_URL, int(self.img_id), i, self.ext))
-            self.downloader.download(download_queue, self.id)
+            for i in range(len(self.ext)):
+                download_queue.append('%s/%d/%d.%s' % (IMAGE_URL, int(self.img_id), i+1, EXT_MAP[self.ext[i]]))
+
+            self.downloader.download(download_queue, format_filename('%s-%s' % (self.id, self.name[:200])))
         else:
-            logger.critical('Downloader has not be loaded')
+            logger.critical('Downloader has not been loaded')
 
 
 if __name__ == '__main__':

nhentai/downloader.py

@@ -14,6 +14,9 @@ from nhentai.parser import request
 from nhentai.utils import Singleton
 
 
+requests.packages.urllib3.disable_warnings()
+
+
 class NhentaiImageNotExistException(Exception):
     pass
 
@@ -29,12 +32,27 @@ class Downloader(Singleton):
         self.timeout = timeout
 
     def _download(self, url, folder='', filename='', retried=0):
-        logger.info('Start downloading: {0} ...'.format(url))
+        logger.info('Starting to download {0} ...'.format(url))
         filename = filename if filename else os.path.basename(urlparse(url).path)
         base_filename, extension = os.path.splitext(filename)
         try:
+            if os.path.exists(os.path.join(folder, base_filename.zfill(3) + extension)):
+                logger.warning('File: {0} exists, ignoring'.format(os.path.join(folder, base_filename.zfill(3) +
+                                                                                extension)))
+                return 1, url
+
             with open(os.path.join(folder, base_filename.zfill(3) + extension), "wb") as f:
+                i=0
+                while i<10:
+                    try:
                         response = request('get', url, stream=True, timeout=self.timeout)
+                    except Exception as e:
+                        i+=1
+                        if not i<10:
+                            logger.critical(str(e))
+                            return 0, None
+                        continue
+                    break
                 if response.status_code != 200:
                     raise NhentaiImageNotExistException
                 length = response.headers.get('content-length')
@@ -44,7 +62,7 @@ class Downloader(Singleton):
                     for chunk in response.iter_content(2048):
                         f.write(chunk)
 
-        except requests.HTTPError as e:
+        except (requests.HTTPError, requests.Timeout) as e:
             if retried < 3:
                 logger.warning('Warning: {0}, retrying({1}) ...'.format(str(e), retried))
                 return 0, self._download(url=url, folder=folder, filename=filename, retried=retried+1)
@@ -64,22 +82,22 @@ class Downloader(Singleton):
     def _download_callback(self, request, result):
         result, data = result
         if result == 0:
-            logger.critical('fatal errors occurred, quit.')
-            exit(1)
+            logger.warning('fatal errors occurred, ignored')
+            # exit(1)
         elif result == -1:
             logger.warning('url {} return status code 404'.format(data))
         else:
-            logger.log(15, '{0} download successfully'.format(data))
+            logger.log(15, '{0} downloaded successfully'.format(data))
 
     def download(self, queue, folder=''):
-        if not isinstance(folder, (text)):
+        if not isinstance(folder, text):
             folder = str(folder)
 
         if self.path:
             folder = os.path.join(self.path, folder)
 
         if not os.path.exists(folder):
-            logger.warn('Path \'{0}\' not exist.'.format(folder))
+            logger.warn('Path \'{0}\' does not exist, creating.'.format(folder))
             try:
                 os.makedirs(folder)
             except EnvironmentError as e:

nhentai/logger.py

@@ -104,6 +104,9 @@ class ColorizingStreamHandler(logging.StreamHandler):
                 text = parts.pop(0)
 
                 if text:
+                    if sys.version_info < (3, 0, 0):
+                        write(text.encode('utf-8'))
+                    else:
                         write(text)
 
                 if parts:

nhentai/parser.py

@@ -1,9 +1,12 @@
 # coding: utf-8
 from __future__ import unicode_literals, print_function
 
-from bs4 import BeautifulSoup
+import os
 import re
+import threadpool
 import requests
+import time
+from bs4 import BeautifulSoup
 from tabulate import tabulate
 
 import nhentai.constant as constant
@@ -17,81 +20,147 @@ def request(method, url, **kwargs):
     return requests.__dict__[method](url, proxies=constant.PROXY, verify=False, **kwargs)
 
 
+def login_parser(username, password):
+    s = requests.Session()
+    s.proxies = constant.PROXY
+    s.verify = False
+    s.headers.update({'Referer': constant.LOGIN_URL})
+
+    s.get(constant.LOGIN_URL)
+    content = s.get(constant.LOGIN_URL).content
+    html = BeautifulSoup(content, 'html.parser')
+    csrf_token_elem = html.find('input', attrs={'name': 'csrfmiddlewaretoken'})
+
+    if not csrf_token_elem:
+        raise Exception('Cannot find csrf token to login')
+    csrf_token = csrf_token_elem.attrs['value']
+
+    login_dict = {
+        'csrfmiddlewaretoken': csrf_token,
+        'username_or_email': username,
+        'password': password,
+    }
+    resp = s.post(constant.LOGIN_URL, data=login_dict)
+    if 'Invalid username/email or password' in resp.text:
+        logger.error('Login failed, please check your username and password')
+        exit(1)
+
+    html = BeautifulSoup(s.get(constant.FAV_URL).content, 'html.parser')
+    count = html.find('span', attrs={'class': 'count'})
+    if not count:
+        logger.error("Can't get your number of favorited doujins. Did the login failed?")
+
+    count = int(count.text.strip('(').strip(')'))
+    if count == 0:
+        logger.warning('No favorites found')
+        return []
+    pages = int(count / 25)
+
+    if pages:
+        pages += 1 if count % (25 * pages) else 0
+    else:
+        pages = 1
+
+    logger.info('You have %d favorites in %d pages.' % (count, pages))
+
+    if os.getenv('DEBUG'):
+        pages = 1
+
+    ret = []
+    doujinshi_id = re.compile('data-id="([\d]+)"')
+
+    def _callback(request, result):
+        ret.append(result)
+
+    thread_pool = threadpool.ThreadPool(5)
+
+    for page in range(1, pages+1):
+        try:
+            logger.info('Getting doujinshi ids of page %d' % page)
+            resp = s.get(constant.FAV_URL + '?page=%d' % page).text
+            ids = doujinshi_id.findall(resp)
+            requests_ = threadpool.makeRequests(doujinshi_parser, ids, _callback)
+            [thread_pool.putRequest(req) for req in requests_]
+            thread_pool.wait()
+        except Exception as e:
+            logger.error('Error: %s, continue', str(e))
+
+    return ret
+
+
 def doujinshi_parser(id_):
     if not isinstance(id_, (int,)) and (isinstance(id_, (str,)) and not id_.isdigit()):
         raise Exception('Doujinshi id({0}) is not valid'.format(id_))
 
     id_ = int(id_)
-    logger.log(15, 'Fetching doujinshi information of id {0}'.format(id_))
+    logger.log(15, 'Fetching information of doujinshi id {0}'.format(id_))
     doujinshi = dict()
     doujinshi['id'] = id_
-    url = '{0}/{1}/'.format(constant.DETAIL_URL, id_)
-
+    url = '{0}/{1}'.format(constant.DETAIL_URL, id_)
+    i=0
+    while i<5:
         try:
-        response = request('get', url).content
+            response = request('get', url).json()
         except Exception as e:
+            i+=1
+            if not i<5:
                 logger.critical(str(e))
                 exit(1)
-
-    html = BeautifulSoup(response, 'html.parser')
-    doujinshi_info = html.find('div', attrs={'id': 'info'})
-
-    title = doujinshi_info.find('h1').text
-    subtitle = doujinshi_info.find('h2')
-
-    doujinshi['name'] = title
-    doujinshi['subtitle'] = subtitle.text if subtitle else ''
-
-    doujinshi_cover = html.find('div', attrs={'id': 'cover'})
-    img_id = re.search('/galleries/([\d]+)/cover\.(jpg|png)$', doujinshi_cover.a.img['src'])
-    if not img_id:
-        logger.critical('Tried yo get image id failed')
-        exit(1)
-
-    doujinshi['img_id'] = img_id.group(1)
-    doujinshi['ext'] = img_id.group(2)
-
-    pages = 0
-    for _ in doujinshi_info.find_all('div', class_=''):
-        pages = re.search('([\d]+) pages', _.text)
-        if pages:
-            pages = pages.group(1)
+            continue
         break
-    doujinshi['pages'] = int(pages)
+
+    doujinshi['name'] = response['title']['english']
+    doujinshi['subtitle'] = response['title']['japanese']
+    doujinshi['img_id'] = response['media_id']
+    doujinshi['ext'] = ''.join(map(lambda s: s['t'], response['images']['pages']))
+    doujinshi['pages'] = len(response['images']['pages'])
 
     # gain information of the doujinshi
-    information_fields = doujinshi_info.find_all('div', attrs={'class': 'field-name'})
-    needed_fields = ['Characters', 'Artists', 'Language', 'Tags']
-    for field in information_fields:
-        field_name = field.contents[0].strip().strip(':')
-        if field_name in needed_fields:
-            data = [sub_field.contents[0].strip() for sub_field in
-                    field.find_all('a', attrs={'class': 'tag'})]
-            doujinshi[field_name.lower()] = ', '.join(data)
+    needed_fields = ['character', 'artist', 'language', 'tag']
+    for tag in response['tags']:
+        tag_type = tag['type']
+        if tag_type in needed_fields:
+            if tag_type == 'tag':
+                if tag_type not in doujinshi:
+                    doujinshi[tag_type] = {}
+
+                tag['name'] = tag['name'].replace(' ', '-')
+                tag['name'] = tag['name'].lower()
+                doujinshi[tag_type][tag['name']] = tag['id']
+            elif tag_type not in doujinshi:
+                doujinshi[tag_type] = tag['name']
+            else:
+                doujinshi[tag_type] += tag['name']
 
     return doujinshi
 
 
 def search_parser(keyword, page):
-    logger.debug('Searching doujinshis of keyword {0}'.format(keyword))
+    logger.debug('Searching doujinshis using keywords {0}'.format(keyword))
     result = []
+    i=0
+    while i<5:
         try:
-        response = request('get', url=constant.SEARCH_URL, params={'q': keyword, 'page': page}).content
-    except requests.ConnectionError as e:
-        logger.critical(e)
+            response = request('get', url=constant.SEARCH_URL, params={'query': keyword, 'page': page}).json()
+        except Exception as e:
+            i+=1
+            if not i<5:
+                logger.critical(str(e))
                 logger.warn('If you are in China, please configure the proxy to fu*k GFW.')
                 exit(1)
+            continue
+        break
+
+    if 'result' not in response:
+        raise Exception('No result in response')
+
+    for row in response['result']:
+        title = row['title']['english']
+        title = title[:85] + '..' if len(title) > 85 else title
+        result.append({'id': row['id'], 'title': title})
 
-    html = BeautifulSoup(response, 'html.parser')
-    doujinshi_search_result = html.find_all('div', attrs={'class': 'gallery'})
-    for doujinshi in doujinshi_search_result:
-        doujinshi_container = doujinshi.find('div', attrs={'class': 'caption'})
-        title = doujinshi_container.text.strip()
-        title = (title[:85] + '..') if len(title) > 85 else title
-        id_ = re.search('/g/(\d+)/', doujinshi.a['href']).group(1)
-        result.append({'id': id_, 'title': title})
     if not result:
-        logger.warn('Not found anything of keyword {}'.format(keyword))
+        logger.warn('No results for keywords {}'.format(keyword))
 
     return result
 
@@ -104,5 +173,86 @@ def print_doujinshi(doujinshi_list):
     logger.info('Search Result\n' +
                 tabulate(tabular_data=doujinshi_list, headers=headers, tablefmt='rst'))
 
+
+def tag_parser(tag_id, max_page=1):
+    logger.info('Searching for doujinshi with tag id {0}'.format(tag_id))
+    result = []
+    i=0
+    while i<5:
+        try:
+            response = request('get', url=constant.TAG_API_URL, params={'sort': 'popular', 'tag_id': tag_id}).json()
+        except Exception as e:
+            i+=1
+            if not i<5:
+                logger.critical(str(e))
+                exit(1)
+            continue
+        break
+    page = max_page if max_page <= response['num_pages'] else int(response['num_pages'])
+
+    for i in range(1, page+1):
+        logger.info('Getting page {} ...'.format(i))
+
+        if page != 1:
+            i=0
+            while i<5:
+                try:
+                    response = request('get', url=constant.TAG_API_URL, params={'sort': 'popular', 'tag_id': tag_id}).json()
+                except Exception as e:
+                    i+=1
+                    if not i<5:
+                        logger.critical(str(e))
+                        exit(1)
+                    continue
+                break
+    for row in response['result']:
+        title = row['title']['english']
+        title = title[:85] + '..' if len(title) > 85 else title
+        result.append({'id': row['id'], 'title': title})
+
+    if not result:
+        logger.warn('No results for tag id {}'.format(tag_id))
+    
+    return result
+
+
+def tag_guessing(tag_name):
+    tag_name = tag_name.lower()
+    tag_name = tag_name.replace(' ', '-')
+    logger.info('Trying to get tag_id of tag \'{0}\''.format(tag_name))
+    i=0
+    while i<5:
+        try:
+            response = request('get', url='%s/%s' % (constant.TAG_URL, tag_name)).content
+        except Exception as e:
+            i+=1
+            if not i<5:
+                logger.critical(str(e))
+                exit(1)
+            continue
+        break
+
+    html = BeautifulSoup(response, 'html.parser')
+    first_item = html.find('div', attrs={'class': 'gallery'})
+    if not first_item:
+        logger.error('Cannot find doujinshi id of tag \'{0}\''.format(tag_name))
+        return
+
+    doujinshi_id = re.findall('(\d+)', first_item.a.attrs['href'])
+    if not doujinshi_id:
+        logger.error('Cannot find doujinshi id of tag \'{0}\''.format(tag_name))
+        return
+
+    ret = doujinshi_parser(doujinshi_id[0])
+    if 'tag' in ret and tag_name in ret['tag']:
+        tag_id = ret['tag'][tag_name]
+        logger.info('Tag id of tag \'{0}\' is {1}'.format(tag_name, tag_id))
+    else:
+        logger.error('Cannot find doujinshi id of tag \'{0}\''.format(tag_name))
+        return
+
+    return tag_id
+
+
 if __name__ == '__main__':
     print(doujinshi_parser("32271"))

nhentai/utils.py

@@ -1,6 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals, print_function
 
+import sys
+import os
+import string
+import zipfile
+import shutil
+from nhentai.logger import logger
+
 
 class _Singleton(type):
     """ A metaclass that creates a Singleton base class when called. """
@@ -23,3 +30,98 @@ def urlparse(url):
         from urllib.parse import urlparse
 
     return urlparse(url)
+
+
+def readfile(path):
+    loc = os.path.dirname(__file__)
+
+    with open(os.path.join(loc, path), 'r') as file:
+        return file.read()
+
+
+def generate_html(output_dir='.', doujinshi_obj=None):
+    image_html = ''
+
+    if doujinshi_obj is not None:
+        doujinshi_dir = os.path.join(output_dir, format_filename('%s-%s' % (doujinshi_obj.id,
+                                                                            doujinshi_obj.name)))
+    else:
+        doujinshi_dir = '.'
+
+    file_list = os.listdir(doujinshi_dir)
+    file_list.sort()
+
+    for image in file_list:
+        if not os.path.splitext(image)[1] in ('.jpg', '.png'):
+            continue
+
+        image_html += '<img src="{0}" class="image-item"/>\n'\
+            .format(image)
+
+    html = readfile('viewer/index.html')
+    css = readfile('viewer/styles.css')
+    js = readfile('viewer/scripts.js')
+
+    if doujinshi_obj is not None:
+        title = doujinshi_obj.name
+        if sys.version_info < (3, 0):
+            title = title.encode('utf-8')
+    else:
+        title = 'nHentai HTML Viewer'
+
+    data = html.format(TITLE=title, IMAGES=image_html, SCRIPTS=js, STYLES=css)
+    try:
+        if sys.version_info < (3, 0):
+            with open(os.path.join(doujinshi_dir, 'index.html'), 'w') as f:
+                f.write(data)
+        else:
+            with open(os.path.join(doujinshi_dir, 'index.html'), 'wb') as f:
+                f.write(data.encode('utf-8'))
+
+        logger.log(15, 'HTML Viewer has been write to \'{0}\''.format(os.path.join(doujinshi_dir, 'index.html')))
+    except Exception as e:
+        logger.warning('Writen HTML Viewer failed ({})'.format(str(e)))
+
+
+def generate_cbz(output_dir='.', doujinshi_obj=None):
+    if doujinshi_obj is not None:
+        doujinshi_dir = os.path.join(output_dir, format_filename('%s-%s' % (doujinshi_obj.id,
+                                                                            str(doujinshi_obj.name[:200]))))    
+        cbz_filename = os.path.join(output_dir, format_filename('%s-%s.cbz' % (doujinshi_obj.id,
+                                                                            str(doujinshi_obj.name[:200]))))
+    else:
+        cbz_filename = './doujinshi.cbz'
+        doujinshi_dir = '.'
+
+    file_list = os.listdir(doujinshi_dir)
+    file_list.sort()
+    
+    with zipfile.ZipFile(cbz_filename, 'w') as cbz_pf:
+        for image in file_list:
+            image_path = os.path.join(doujinshi_dir, image)
+            cbz_pf.write(image_path, image)
+            
+    shutil.rmtree(doujinshi_dir, ignore_errors=True)
+    logger.log(15, 'Comic Book CBZ file has been write to \'{0}\''.format(doujinshi_dir))
+
+
+
+
+
+
+
+def format_filename(s):
+    """Take a string and return a valid filename constructed from the string.
+Uses a whitelist approach: any characters not present in valid_chars are
+removed. Also spaces are replaced with underscores.
+
+Note: this method may produce invalid filenames such as ``, `.` or `..`
+When I use this method I prepend a date string like '2009_01_15_19_46_32_'
+and append a file extension like '.txt', so I avoid the potential of using
+an invalid filename.
+
+"""
+    valid_chars = "-_.() %s%s" % (string.ascii_letters, string.digits)
+    filename = ''.join(c for c in s if c in valid_chars)
+    filename = filename.replace(' ', '_')  # I don't like spaces in filenames.
+    return filename

nhentai/viewer/index.html

@@ -0,0 +1,24 @@
+<!DOCTYPE html>
+<html>
+<head>
+    <meta charset="UTF-8">
+    <title>{TITLE}</title>
+    <style>
+{STYLES}
+    </style>
+</head>
+<body>
+
+<nav id="list">
+{IMAGES}</nav>
+
+<div id="image-container">
+    <span id="page-num"></span>
+    <div id="dest"></div>
+</div>
+
+<script>
+{SCRIPTS}
+</script>
+</body>
+</html>

nhentai/viewer/scripts.js

@@ -0,0 +1,62 @@
+const pages = Array.from(document.querySelectorAll('img.image-item'));
+let currentPage = 0;
+
+function changePage(pageNum) {
+    const previous = pages[currentPage];
+    const current = pages[pageNum];
+
+    if (current == null) {
+        return;
+    }
+    
+    previous.classList.remove('current');
+    current.classList.add('current');
+
+    currentPage = pageNum;
+
+    const display = document.getElementById('dest');
+    display.style.backgroundImage = `url("${current.src}")`;
+
+    document.getElementById('page-num')
+        .innerText = [
+                (pageNum + 1).toLocaleString(),
+                pages.length.toLocaleString()
+            ].join('\u200a/\u200a');
+}
+
+changePage(0);
+
+document.getElementById('list').onclick = event => {
+    if (pages.includes(event.target)) {
+        changePage(pages.indexOf(event.target));
+    }
+};
+
+document.getElementById('image-container').onclick = event => {
+    const width = document.getElementById('image-container').clientWidth;
+    const clickPos = event.clientX / width;
+
+    if (clickPos < 0.5) {
+        changePage(currentPage - 1);
+    } else {
+        changePage(currentPage + 1);
+    }
+};
+
+document.onkeypress = event => {
+    switch (event.key.toLowerCase()) {
+        // Previous Image
+        case 'arrowleft':
+        case 'a':
+            changePage(currentPage - 1);
+            break;
+
+        // Next Image
+        case ' ':
+        case 'enter':
+        case 'arrowright':
+        case 'd':
+            changePage(currentPage + 1);
+            break;
+    }
+};

nhentai/viewer/styles.css

@@ -0,0 +1,69 @@
+*, *::after, *::before {
+    box-sizing: border-box;
+}
+
+img {
+    vertical-align: middle;
+}
+
+html, body {
+    display: flex;
+    background-color: #e8e6e6;
+    height: 100%;
+    width: 100%;
+    padding: 0;
+    margin: 0;
+    font-family: sans-serif;
+}
+
+#list {
+    height: 100%;
+    overflow: auto;
+    width: 260px;
+    text-align: center;
+}
+
+#list img {
+    width: 200px;
+    padding: 10px;
+    border-radius: 10px;
+    margin: 15px 0;
+    cursor: pointer;
+}
+
+#list img.current {
+    background: #0003;
+}
+
+#image-container {
+    flex: auto;
+    height: 100vh;
+    background: #222;
+    color: #fff;
+    text-align: center;
+    cursor: pointer;
+    -webkit-user-select: none;
+    user-select: none;
+    position: relative;
+}
+
+#image-container #dest {
+    height: 100%;
+    width: 100%;
+    background-size: contain;
+    background-repeat: no-repeat;
+    background-position: center;
+}
+
+#image-container #page-num {
+    position: absolute;
+    font-size: 18pt;
+    left: 10px;
+    bottom: 5px;
+    font-weight: bold;
+    opacity: 0.75;
+    text-shadow: /* Duplicate the same shadow to make it very strong */
+        0 0 2px #222,
+        0 0 2px #222,
+        0 0 2px #222;
+}

requirements.txt

@@ -2,4 +2,4 @@ requests>=2.5.0
 BeautifulSoup4>=4.0.0
 threadpool>=1.2.7
 tabulate>=0.7.5
-future>=0.15.2
+future>=0.15.2threadpool==1.3.2

setup.py

@@ -1,15 +1,18 @@
 # coding: utf-8
 from __future__ import print_function, unicode_literals
+import sys
 import codecs
 from setuptools import setup, find_packages
 from nhentai import __version__, __author__, __email__
 
+
 with open('requirements.txt') as f:
     requirements = [l for l in f.read().splitlines() if l]
 
 
 def long_description():
     with codecs.open('README.md', 'rb') as f:
+        if sys.version_info >= (3, 0, 0):
             return str(f.read())
 
 setup(