update user-agent

download ids from file

.travis.yml

@@ -12,8 +12,10 @@ install:
     - python setup.py install
 
 script:
+    - echo 268642 > /tmp/test.txt
     - 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 --download --output=/tmp/
     - NHENTAI=https://nhentai.net nhentai --tag lolicon
     - NHENTAI=https://nhentai.net nhentai --id 92066 --output=/tmp/ --cbz
+    - NHENTAI=https://nhentai.net nhentai --file /tmp/test.txt

README.md

@@ -18,17 +18,24 @@ nHentai is a CLI tool for downloading doujinshi from [nhentai.net](http://nhenta
     cd nhentai
     python setup.py install
     
-### Gentoo
+### Installation (Gentoo)
 
     layman -fa glicOne
     sudo emerge net-misc/nhentai
 
 ### Usage
+**IMPORTANT**: To bypass the nhentai frequency limit, you should use `--login` option to log into nhentai.net.
+
 Download specified doujinshi:
 ```bash
 nhentai --id=123855,123866
 ```
 
+Download doujinshi with ids specified in a file:
+```bash
+nhentai --file=doujinshi.txt
+```
+
 Search a keyword and download the first page:
 ```bash
 nhentai --search="tomori" --page=1 --download

doujinshi.txt

@@ -0,0 +1,5 @@
+184212
+204944
+222460
+244502
+261909

nhentai/__init__.py

@@ -1,3 +1,3 @@
-__version__ = '0.2.19'
+__version__ = '0.3.1'
 __author__ = 'RicterZ'
 __email__ = 'ricterzheng@gmail.com'

nhentai/cmdline.py

@@ -37,6 +37,7 @@ def banner():
 def cmd_parser():
     parser = OptionParser('\n  nhentai --search [keyword] --download'
                           '\n  NHENTAI=http://h.loli.club nhentai --id [ID ...]'
+                          '\n  nhentai --file [filename]'    
                           '\n\nEnvironment Variable:\n'
                           '  NHENTAI                 nhentai mirror url')
     parser.add_option('--download', dest='is_download', action='store_true',
@@ -70,6 +71,7 @@ def cmd_parser():
                       help='Generate Comic Book CBZ File')
     parser.add_option('--rm-origin-dir', dest='rm_origin_dir', action='store_true', default=False,
                       help='Remove downloaded doujinshi dir when generated CBZ file.')
+    parser.add_option('--file',  '-f', type='string', dest='file', action='store', help='Read gallery IDs from file.')
 
     try:
         sys.argv = list(map(lambda x: unicode(x.decode(sys.stdin.encoding)), sys.argv))
@@ -98,6 +100,11 @@ def cmd_parser():
         _ = map(lambda id: id.strip(), args.id.split(','))
         args.id = set(map(int, filter(lambda id_: id_.isdigit(), _)))
 
+    if args.file:
+        with open(args.file, 'r') as f:
+            _ = map(lambda id: id.strip(), f.readlines())
+            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 and \
             not args.login and not args.tag:
         logger.critical('Doujinshi id(s) are required for downloading')

nhentai/command.py

@@ -5,7 +5,7 @@ import signal
 import platform
 
 from nhentai.cmdline import cmd_parser, banner
-from nhentai.parser import doujinshi_parser, search_parser, print_doujinshi, login_parser, tag_guessing, tag_parser
+from nhentai.parser import doujinshi_parser, search_parser, print_doujinshi, login_parser, tag_parser, login
 from nhentai.doujinshi import Doujinshi
 from nhentai.downloader import Downloader
 from nhentai.logger import logger
@@ -24,13 +24,14 @@ def main():
     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):
+        login(username, password)
+
+        if options.is_download:
+            for doujinshi_info in login_parser():
                 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)
+        doujinshis = tag_parser(options.tag, max_page=options.max_page)
         print_doujinshi(doujinshis)
         if options.is_download:
             doujinshi_ids = map(lambda d: d['id'], doujinshis)

nhentai/constant.py

@@ -5,11 +5,16 @@ from nhentai.utils import urlparse
 
 BASE_URL = os.getenv('NHENTAI', 'https://nhentai.net')
 
-DETAIL_URL = '%s/api/gallery' % BASE_URL
-SEARCH_URL = '%s/api/galleries/search' % BASE_URL
+__api_suspended_DETAIL_URL = '%s/api/gallery' % BASE_URL
+__api_suspended_SEARCH_URL = '%s/api/galleries/search' % BASE_URL
+
+DETAIL_URL = '%s/g' % BASE_URL
+SEARCH_URL = '%s/search/' % BASE_URL
+
 TAG_URL = '%s/tag' % BASE_URL
 TAG_API_URL = '%s/api/galleries/tagged' % BASE_URL
 LOGIN_URL = '%s/login/' % BASE_URL
+CHALLENGE_URL = '%s/challenge' % BASE_URL
 FAV_URL = '%s/favorites/' % BASE_URL
 
 u = urlparse(BASE_URL)

nhentai/doujinshi.py

@@ -58,9 +58,20 @@ class Doujinshi(object):
         logger.info('Starting to download doujinshi: %s' % self.name)
         if self.downloader:
             download_queue = []
+
+            if len(self.ext) != self.pages:
+                logger.warning('Page count and ext count do not equal')
+
+            for i in range(1, min(self.pages, len(self.ext)) + 1):
+                download_queue.append('%s/%d/%d.%s' % (IMAGE_URL, int(self.img_id), i, self.ext[i-1]))
+
+            self.downloader.download(download_queue, self.filename)
+
+            '''
             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, self.filename)
+            '''
+
         else:
             logger.critical('Downloader has not been loaded')
 

nhentai/parser.py

@@ -13,42 +13,64 @@ import nhentai.constant as constant
 from nhentai.logger import logger
 
 
+session = requests.Session()
+session.headers.update({
+    'Referer': constant.LOGIN_URL,
+    'User-Agent': 'nhentai command line client (https://github.com/RicterZ/nhentai)',
+})
+
+
 def request(method, url, **kwargs):
-    if not hasattr(requests, method):
-        raise AttributeError('\'requests\' object has no attribute \'{0}\''.format(method))
+    global session
+    if not hasattr(session, method):
+        raise AttributeError('\'requests.Session\' object has no attribute \'{0}\''.format(method))
 
-    return requests.__dict__[method](url, proxies=constant.PROXY, verify=False, **kwargs)
+    return getattr(session, 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
+def _get_csrf_token(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']
+    return csrf_token_elem.attrs['value']
+
+
+def login(username, password):
+    csrf_token = _get_csrf_token(request('get', url=constant.LOGIN_URL).text)
+    if os.getenv('DEBUG'):
+        logger.info('Getting CSRF token ...')
+
+    if os.getenv('DEBUG'):
+        logger.info('CSRF token is {}'.format(csrf_token))
 
     login_dict = {
         'csrfmiddlewaretoken': csrf_token,
         'username_or_email': username,
         'password': password,
     }
-    resp = s.post(constant.LOGIN_URL, data=login_dict)
+    resp = request('post', url=constant.LOGIN_URL, data=login_dict)
+
+    if 'You\'re loading pages way too quickly.' in resp.text:
+        csrf_token = _get_csrf_token(resp.text)
+        resp = request('post', url=resp.url, data={'csrfmiddlewaretoken': csrf_token, 'next': '/'})
+
     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')
+    if 'You\'re loading pages way too quickly.' in resp.text:
+        logger.error('You meet challenge insistently, please submit a issue'
+                     ' at https://github.com/RicterZ/nhentai/issues')
+        exit(2)
+
+
+def login_parser():
+    html = BeautifulSoup(request('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?")
+        return
 
     count = int(count.text.strip('(').strip(')').replace(',', ''))
     if count == 0:
@@ -77,7 +99,7 @@ def login_parser(username, password):
     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
+            resp = request('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_]
@@ -92,13 +114,98 @@ 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_))
+    doujinshi = dict()
+    doujinshi['id'] = id_
+    url = '{0}/{1}/'.format(constant.DETAIL_URL, id_)
+
+    try:
+        response = request('get', url).content
+    except Exception as e:
+        logger.critical(str(e))
+        raise SystemExit
+
+    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.attrs['data-src'])
+
+    ext = []
+    for i in html.find_all('div', attrs={'class': 'thumb-container'}):
+        _, ext_name = os.path.basename(i.img.attrs['data-src']).rsplit('.', 1)
+        ext.append(ext_name)
+
+    if not img_id:
+        logger.critical('Tried yo get image id failed')
+        exit(1)
+
+    doujinshi['img_id'] = img_id.group(1)
+    doujinshi['ext'] = ext
+
+    pages = 0
+    for _ in doujinshi_info.find_all('div', class_=''):
+        pages = re.search('([\d]+) pages', _.text)
+        if pages:
+            pages = pages.group(1)
+            break
+    doujinshi['pages'] = int(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)
+
+    return doujinshi
+
+
+def search_parser(keyword, page):
+    logger.debug('Searching doujinshis of keyword {0}'.format(keyword))
+    result = []
+    try:
+        response = request('get', url=constant.SEARCH_URL, params={'q': keyword, 'page': page}).content
+    except requests.ConnectionError as e:
+        logger.critical(e)
+        logger.warn('If you are in China, please configure the proxy to fu*k GFW.')
+        raise SystemExit
+
+    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 if len(title) < 85 else title[:82] + '...'
+        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))
+
+    return result
+
+
+def __api_suspended_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 information of doujinshi id {0}'.format(id_))
     doujinshi = dict()
     doujinshi['id'] = id_
     url = '{0}/{1}'.format(constant.DETAIL_URL, id_)
     i = 0
-    while i<5:
+    while 5 > i:
         try:
             response = request('get', url).json()
         except Exception as e:
@@ -135,7 +242,7 @@ def doujinshi_parser(id_):
     return doujinshi
 
 
-def search_parser(keyword, page):
+def __api_suspended_search_parser(keyword, page):
     logger.debug('Searching doujinshis using keywords {0}'.format(keyword))
     result = []
     i = 0
@@ -174,37 +281,18 @@ def print_doujinshi(doujinshi_list):
                 tabulate(tabular_data=doujinshi_list, headers=headers, tablefmt='rst'))
 
 
-def tag_parser(tag_id, max_page=1):
+def __api_suspended_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
+            response = request('get', url=constant.TAG_API_URL,
+                               params={'sort': 'popular', 'tag_id': tag_id}).json()
     for row in response['result']:
         title = row['title']['english']
         title = title[:85] + '..' if len(title) > 85 else title
@@ -216,42 +304,31 @@ def tag_parser(tag_id, max_page=1):
     return result
 
 
-def tag_guessing(tag_name):
+def tag_parser(tag_name, max_page=1):
+    result = []
     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
+
+    for p in range(1, max_page + 1):
+        logger.debug('Fetching page {0} for doujinshi with tag \'{1}\''.format(p, tag_name))
+        response = request('get', url='%s/%s?page=%d' % (constant.TAG_URL, tag_name, p)).content
 
         html = BeautifulSoup(response, 'html.parser')
-    first_item = html.find('div', attrs={'class': 'gallery'})
-    if not first_item:
+        doujinshi_items = html.find_all('div', attrs={'class': 'gallery'})
+        if not doujinshi_items:
             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
+        for i in doujinshi_items:
+            doujinshi_id = i.a.attrs['href'].strip('/g')
+            doujinshi_title = i.a.text.strip()
+            doujinshi_title = doujinshi_title if len(doujinshi_title) < 85 else doujinshi_title[:82] + '...'
+            result.append({'title': doujinshi_title, 'id': doujinshi_id})
 
-    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
+    if not result:
+        logger.warn('No results for tag \'{}\''.format(tag_name))
 
-    return tag_id
+    return result
 
 
 if __name__ == '__main__':

nhentai/utils.py

@@ -121,4 +121,7 @@ an invalid filename.
     filename = filename.replace(' ', '_')  # I don't like spaces in filenames.
     if len(filename) > 100:
         filename = filename[:100] + '...]'
+
+    # Remove [] from filename
+    filename = filename.replace('[]', '')
     return filename